{ "schemaVersion": 1, "kind": "model-index", "count": 234, "generatedAt": "2026-10-07T00:00:00.000Z", "scoreNote": "This index lists published specs, pricing, and source dates. It does not include intelligence, composite, or other derived scores — cite the dated rows on each model page.", "models": [ { "slug": "a-x-k2", "name": "A.X-K2", "provider": "Other", "family": "A.X", "summary": { "en": "SK Telecom's A.X-K2 open agentic MoE. AA lists $0 / $0 per 1M tokens.", "de": "SK Telecoms offenes agentisches MoE A.X-K2. AA führt $0 / $0 pro 1M Tokens." }, "role": { "en": "Korean sovereign-model lane on Artificial Analysis. Prefer sourced ledger rows over launch claims.", "de": "Koreanische Sovereign-Model-Spur auf Artificial Analysis. Bevorzuge belegte Ledger-Zeilen gegenüber Launch-Claims." }, "releaseDate": "2026-08-12", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingNote": { "en": "AA lists $0 / $0 per 1M tokens on the 2026-08-15 snapshot. Context cap is unlisted on the AA record.", "de": "AA führt $0 / $0 pro 1M Tokens im Snapshot vom 2026-08-15. Kontextlimit ist auf dem AA-Nachweis nicht angegeben." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis A.X-K2", "de": "Artificial-Analysis-A.X-K2" }, "url": "https://artificialanalysis.ai/models/a-x-k2", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/a-x-k2" }, { "slug": "agnes-2-5-pro-alpha", "name": "Agnes 2.5 Pro Alpha", "provider": "Other", "family": "Agnes 2.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.45/$0.9 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.45/$0.9 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.45, "outputPriceUSD": 0.9, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/agnes-2-5-pro-alpha" }, { "slug": "agnes-2-5-pro-beta", "name": "Agnes 2.5 Pro Beta", "provider": "Other", "family": "Agnes 2.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.1/$0.3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.1/$0.3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-26", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.1, "outputPriceUSD": 0.3, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/agnes-2-5-pro-beta" }, { "slug": "apodex-1-1", "name": "Apodex 1.1", "provider": "Other", "family": "Apodex 1.1", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-30", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 3, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/apodex-1-1" }, { "slug": "atria-dawn-preview", "name": "Atria Dawn Preview", "provider": "Other", "family": "Atria", "summary": { "en": "Shanghai AI Laboratory's MIT-licensed agentic preview, built on the 744B-parameter MoE GLM-5.2 foundation. Text-only. 256K context, 65,536 max output. Hosted at api.atria-asi.ai as Atria-Dawn-Preview; no public per-token list as of 20 Sep 2026.", "de": "MIT-lizenzierte agentische Preview des Shanghai AI Laboratory, auf der 744B-Parameter-MoE-Grundlage GLM-5.2. Nur Text. 256K-Kontext, 65.536 max. Gehostet unter api.atria-asi.ai als Atria-Dawn-Preview; keine öffentliche Per-Token-Liste Stand 20. Sep. 2026." }, "role": { "en": "Open-weight research and engineering agent for plan, code, experiment, and retry loops — not a multimodal chat model.", "de": "Open-Weight-Research- und Engineering-Agent für Plan-, Code-, Experiment- und Retry-Schleifen — kein multimodales Chat-Modell." }, "releaseDate": "2026-09-11", "knowledgeCutoff": "unpublished", "contextWindow": 256000, "maxOutput": 65536, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "MIT weights on Hugging Face (internlm/Atria-Dawn-Preview and FP8) and ModelScope. Hosted international API at https://api.atria-asi.ai/v1 (China: discovery.intern-ai.org.cn). Model ID Atria-Dawn-Preview is case-sensitive. Output cap 1–65,536 tokens. Official API docs publish no per-million-token list; do not treat third-party gateway $0 rows as a vendor price. Self-host via SGLang v0.5.13.post1+ or vLLM v0.23.0+.", "de": "MIT-Weights auf Hugging Face (internlm/Atria-Dawn-Preview und FP8) und ModelScope. Internationale Hosted-API unter https://api.atria-asi.ai/v1 (China: discovery.intern-ai.org.cn). Modell-ID Atria-Dawn-Preview ist case-sensitive. Output-Limit 1–65.536 Tokens. Die offiziellen API-Docs nennen keine Per-Million-Token-Liste; Gateway-Zeilen mit $0 sind kein Vendor-Preis. Self-Host über SGLang v0.5.13.post1+ oder vLLM v0.23.0+." }, "pricingChecked": "2026-09-20", "status": "solid", "sources": [ { "label": { "en": "Hugging Face Atria Dawn Preview", "de": "Hugging Face Atria Dawn Preview" }, "url": "https://huggingface.co/internlm/Atria-Dawn-Preview", "checkedAt": "2026-09-20" }, { "label": { "en": "Atria Dawn Preview API docs", "de": "Atria-Dawn-Preview-API-Docs" }, "url": "https://api.atria-asi.ai/docs", "checkedAt": "2026-09-20" }, { "label": { "en": "Atria Dawn technical report (arXiv:2609.15818)", "de": "Atria-Dawn-Technikbericht (arXiv:2609.15818)" }, "url": "https://arxiv.org/abs/2609.15818", "checkedAt": "2026-09-20" } ], "html": "https://verdictpal.com/models/atria-dawn-preview" }, { "slug": "claude-3-5-sonnet-2024-06", "name": "Claude 3.5 Sonnet (2024-06)", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "June 2024 Claude 3.5 Sonnet checkpoint. Baseline for comparing Sonnet 4.x gains.", "de": "Claude-3.5-Sonnet-Checkpoint Juni 2024. Baseline zum Vergleich der Sonnet-4.x-Gewinne." }, "role": { "en": "Historical Sonnet baseline on public leaderboards — not a current deployment pick.", "de": "Historische Sonnet-Baseline auf öffentlichen Leaderboards — keine aktuelle Deployment-Wahl." }, "releaseDate": "2024-06-20", "knowledgeCutoff": "2024-04-01", "contextWindow": 200000, "maxOutput": 8192, "modalities": [ "text", "image", "code", "tool-use" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Claude 3.5 Sonnet", "de": "Claude 3.5 Sonnet" }, "url": "https://www.anthropic.com/news/claude-3-5-sonnet", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/claude-3-5-sonnet-2024-06" }, { "slug": "claude-3-7-sonnet-2025-02", "name": "Claude 3.7 Sonnet", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "Anthropic's February 2025 Sonnet with extended thinking. Superseded by Sonnet 4.6 but still referenced on legacy leaderboards.", "de": "Anthropics Sonnet Februar 2025 mit Extended Thinking. Von Sonnet 4.6 abgelöst, aber weiterhin auf Legacy-Leaderboards referenziert." }, "role": { "en": "Legacy Anthropic mid-tier — migrate new work to Claude Sonnet 4.6.", "de": "Legacy-Anthropic-Mittelklasse — neue Arbeit auf Claude Sonnet 4.6 migrieren." }, "releaseDate": "2025-02-19", "knowledgeCutoff": "2024-10-01", "contextWindow": 200000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Claude 3.7 Sonnet announcement", "de": "Claude-3.7-Sonnet-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-3-7-sonnet", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/claude-3-7-sonnet-2025-02" }, { "slug": "claude-3-opus", "name": "Claude 3 Opus", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's March 2024 Opus.", "de": "Anthropics Opus März 2024." }, "role": { "en": "Pre-Opus-4 historical baseline — not a current pick.", "de": "Historische Baseline vor Opus 4 — keine aktuelle Wahl." }, "releaseDate": "2024-03-04", "knowledgeCutoff": "2023-08-01", "contextWindow": 200000, "maxOutput": 4096, "modalities": [ "text", "image", "code", "tool-use" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 15, "outputPriceUSD": 75, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Claude 3 family", "de": "Claude-3-Familie" }, "url": "https://www.anthropic.com/news/claude-3-family", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/claude-3-opus" }, { "slug": "claude-4-1-opus-thinking", "name": "Claude 4.1 Opus (Reasoning)", "provider": "Anthropic", "family": "Claude 4.1", "summary": { "en": "Listed on Artificial Analysis with API pricing $15/$75 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $15/$75 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-08-05", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 15, "outputPriceUSD": 75, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/claude-4-1-opus-thinking" }, { "slug": "claude-4-5-sonnet-thinking", "name": "Claude 4.5 Sonnet (Reasoning)", "provider": "Anthropic", "family": "Claude 4.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $3/$15 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-09-29", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/claude-4-5-sonnet-thinking" }, { "slug": "claude-4-opus-thinking", "name": "Claude 4 Opus (Reasoning)", "provider": "Anthropic", "family": "Claude 4", "summary": { "en": "Listed on Artificial Analysis with API pricing $15/$75 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $15/$75 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-05-22", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 15, "outputPriceUSD": 75, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/claude-4-opus-thinking" }, { "slug": "claude-fable-5-1", "name": "Claude Fable 5.1", "provider": "Anthropic", "family": "Claude Fable", "summary": { "en": "Anthropic's 1 Sep 2026 Mythos-class refresh of Fable 5. Same $10 / $50 per 1M tokens, cache reads $0.25 per 1M. 1M context, 128K output.", "de": "Anthropics Mythos-class-Refresh von Fable 5 am 1. Sep. 2026. Dieselben $10 / $50 pro 1M Tokens, Cache-Reads $0,25 pro 1M. 1M-Kontext, 128K-Output." }, "role": { "en": "Long-horizon coding, multistep research, and dense document work when Opus 5 at higher effort still falls short. Prefer sourced ledger rows over vendor Terminal-Bench-Science tables.", "de": "Long-Horizon-Coding, mehrstufige Recherche und dichte Dokumentarbeit, wenn Opus 5 bei höherem Effort noch nicht reicht. Bevorzuge belegte Ledger-Zeilen gegenüber Vendor-Terminal-Bench-Science-Tabellen." }, "releaseDate": "2026-09-01", "knowledgeCutoff": "2026-06-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 10, "outputPriceUSD": 50, "pricingNote": { "en": "Same input/output list as Fable 5 ($10 / $50 per 1M tokens). Cache reads $0.25 per 1M (0.025× input). Anthropic estimates ~25% lower typical-workload cost vs Fable 5 from cache-read pricing. US-only inference is 1.1×. Model ID claude-fable-5-1.", "de": "Gleiche Input/Output-Liste wie Fable 5 ($10 / $50 pro 1M Tokens). Cache-Reads $0,25 pro 1M (0,025× Input). Anthropic schätzt ~25 % niedrigere Typical-Workload-Kosten vs. Fable 5 durch Cache-Read-Preise. US-only-Inferenz ist 1,1×. Modell-ID claude-fable-5-1." }, "pricingChecked": "2026-09-18", "status": "solid", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/claude-fable-and-mythos-5-1", "checkedAt": "2026-09-18" }, { "label": { "en": "Claude Fable 5.1 product page", "de": "Claude-Fable-5.1-Produktseite" }, "url": "https://www.anthropic.com/claude/fable", "checkedAt": "2026-09-18" }, { "label": { "en": "Anthropic pricing", "de": "Anthropic-Preise" }, "url": "https://www.anthropic.com/pricing", "checkedAt": "2026-09-18" }, { "label": { "en": "Claude API Fable 5.1 overview", "de": "Claude-API-Fable-5.1-Übersicht" }, "url": "https://platform.claude.com/docs/en/models/fable-5-1/overview", "checkedAt": "2026-09-01" }, { "label": { "en": "Vals model page", "de": "Vals-Modellseite" }, "url": "https://www.vals.ai/models/anthropic_claude-fable-5-1", "checkedAt": "2026-09-05" }, { "label": { "en": "Artificial Analysis Claude Fable 5.1", "de": "Artificial-Analysis-Claude-Fable-5.1" }, "url": "https://artificialanalysis.ai/models/claude-fable-5-1", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/claude-fable-5-1" }, { "slug": "claude-fable-5", "name": "Claude Fable 5", "provider": "Anthropic", "family": "Claude Fable", "summary": { "en": "Anthropic's June 2026 Mythos-class GA release. 1M context, 128K output, safety classifiers route high-risk queries to Opus 4.8. $10 / $50 per 1M tokens.", "de": "Anthropics Mythos-class-GA-Release Juni 2026. 1M-Kontext, 128K-Output, Safety-Klassifikatoren leiten Hochrisiko-Anfragen an Opus 4.8. $10 / $50 pro 1M Tokens." }, "role": { "en": "Long-horizon agentic work — multi-day coding migrations, finance analysis, vision-heavy rebuilds, and research pipelines that outlast Sonnet or Opus sessions.", "de": "Langfristige Agent-Arbeit — mehrtägige Coding-Migrationen, Finance-Analyse, vision-lastige Rebuilds und Research-Pipelines, die Sonnet- oder Opus-Sessions überdauern." }, "releaseDate": "2026-06-09", "knowledgeCutoff": "2026-04-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 10, "outputPriceUSD": 50, "pricingNote": { "en": "Less than half the list price of Claude Mythos Preview. Subscription access rolls out in stages; API and consumption-based Enterprise plans are GA from launch.", "de": "Weniger als die Hälfte des Listenpreises von Claude Mythos Preview. Abo-Zugang rollt stufenweise aus; API und verbrauchsbasierte Enterprise-Pläne sind ab Launch GA." }, "pricingChecked": "2026-06-09", "status": "flagship", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-fable-5-mythos-5", "checkedAt": "2026-06-09" }, { "label": { "en": "Anthropic pricing", "de": "Anthropic-Preise" }, "url": "https://www.anthropic.com/pricing", "checkedAt": "2026-06-09" }, { "label": { "en": "Vals model page", "de": "Vals-Modellseite" }, "url": "https://www.vals.ai/models/anthropic_claude-fable-5", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/claude-fable-5" }, { "slug": "claude-haiku-4-5", "name": "Claude Haiku 4.5", "provider": "Anthropic", "family": "Claude Haiku", "summary": { "en": "Anthropic's speed tier. 200K context, 64K output, $1/$5 per 1M tokens. Best for high-volume, low-latency tasks.", "de": "Anthropics Speed-Stufe. 200K-Kontext, 64K-Output, $1/$5 pro 1M Tokens. Ideal für hochvolumige, latenzarme Aufgaben." }, "role": { "en": "Fast, cheap Claude for routing, classification, extraction, and high-volume backend tasks.", "de": "Schnelles, günstiges Claude für Routing, Klassifikation, Extraktion und hochvolumige Backend-Aufgaben." }, "releaseDate": "2025-10-01", "knowledgeCutoff": "2025-08-01", "contextWindow": 200000, "maxOutput": 64000, "modalities": [ "text", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 1, "outputPriceUSD": 5, "pricingNote": { "en": "5x cheaper than Opus/Sonnet. Prompt caching saves up to 90%.", "de": "5× billiger als Opus/Sonnet. Prompt-Caching spart bis zu 90 %." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "Anthropic model docs", "de": "Anthropic-Modell-Dokumentation" }, "url": "https://www.anthropic.com/docs/about-claude/models", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/claude-haiku-4-5" }, { "slug": "claude-haiku-5-5", "name": "Claude Haiku 5.5", "provider": "Anthropic", "family": "Claude Haiku", "summary": { "en": "Anthropic's small model, released 7 Oct 2026. API id claude-haiku-5-5. $0.10 / $0.50 per 1M tokens on prompts up to 100k; above that, $0.50 / $2.50. First Haiku with an adjustable effort setting.", "de": "Anthropics kleines Modell, veröffentlicht am 7. Okt. 2026. API-ID claude-haiku-5-5. 0,10 $ / 0,50 $ pro 1M Tokens bei Prompts bis 100.000; darüber 0,50 $ / 2,50 $. Erstes Haiku mit einstellbarem Effort." }, "role": { "en": "High-volume work: summaries, compaction, classification, and subagent tasks beside a larger Claude.", "de": "Arbeit mit hohem Volumen: Zusammenfassungen, Kürzen langer Kontexte, Klassifikation und Subagenten neben einem grösseren Claude." }, "releaseDate": "2026-10-07", "knowledgeCutoff": "open", "contextWindow": null, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.1, "outputPriceUSD": 0.5, "pricingNote": { "en": "List price on prompts up to 100k tokens: $0.10 input / $0.50 output per 1M. Over 100k: $0.50 / $2.50. Cache reads $0.01 / $0.05, cache writes $0.125 / $0.625 (short prompt / long prompt). Context window and knowledge cutoff were not on the launch page.", "de": "Listenpreis bei Prompts bis 100.000 Tokens: 0,10 $ Input / 0,50 $ Output pro 1M. Darüber: 0,50 $ / 2,50 $. Cache-Reads 0,01 $ / 0,05 $, Cache-Writes 0,125 $ / 0,625 $ (kurzer Prompt / langer Prompt). Kontextfenster und Knowledge-Cutoff standen nicht auf der Launch-Seite." }, "pricingChecked": "2026-10-07", "status": "solid", "sources": [ { "label": { "en": "Anthropic launch page", "de": "Anthropic-Launchseite" }, "url": "https://www.anthropic.com/claude-haiku-5-5", "checkedAt": "2026-10-07" } ], "html": "https://verdictpal.com/models/claude-haiku-5-5" }, { "slug": "claude-mythos-5-1", "name": "Claude Mythos 5.1", "provider": "Anthropic", "family": "Claude Mythos", "summary": { "en": "Same underlying model as Claude Fable 5.1 with safeguards set for trusted cyberdefense and life-science partners. Project Glasswing / trusted-access only — not general consumer availability. $10 / $50 per 1M tokens; cache reads $0.25.", "de": "Gleiches Basis-Modell wie Claude Fable 5.1 mit Safeguards für vertrauenswürdige Cyberdefense- und Life-Science-Partner. Nur Project Glasswing / Trusted Access — keine allgemeine Consumer-Verfügbarkeit. $10 / $50 pro 1M Tokens; Cache-Reads $0,25." }, "role": { "en": "Restricted Mythos-tier access for cyber defenders and approved life-science researchers — not a default deployment model.", "de": "Eingeschränkter Mythos-Tier-Zugang für Cyberdefender und genehmigte Life-Science-Forscher — kein Standard-Deployment-Modell." }, "releaseDate": "2026-09-01", "knowledgeCutoff": "2026-06-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 10, "outputPriceUSD": 50, "pricingNote": { "en": "Same list pricing as Fable 5.1. Access requires Anthropic trusted-access / Project Glasswing approval. Model ID claude-mythos-5-1.", "de": "Gleicher Listenpreis wie Fable 5.1. Zugang erfordert Anthropic-Trusted-Access- / Project-Glasswing-Freigabe. Modell-ID claude-mythos-5-1." }, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/claude-fable-and-mythos-5-1", "checkedAt": "2026-09-01" }, { "label": { "en": "Fable 5.1 and Mythos 5.1 system card", "de": "System Card Fable 5.1 und Mythos 5.1" }, "url": "https://www-cdn.anthropic.com/0339e6a7c5c7b87f5c07798616dc32c215d14235/Claude%20Fable%205.1%20%26%20Claude%20Mythos%205.1%20System%20Card.pdf", "checkedAt": "2026-09-12" } ], "html": "https://verdictpal.com/models/claude-mythos-5-1" }, { "slug": "claude-mythos-5", "name": "Claude Mythos 5", "provider": "Anthropic", "family": "Claude Mythos", "summary": { "en": "Same underlying model as Claude Fable 5 with safeguards lifted for trusted cyberdefense and research partners. Project Glasswing and a broader trusted-access program — not general consumer availability. $10 / $50 per 1M tokens.", "de": "Gleiches Basis-Modell wie Claude Fable 5 mit gelockerten Safeguards für vertrauenswürdige Cyberdefense- und Research-Partner. Project Glasswing und ein breiteres Trusted-Access-Programm — keine allgemeine Consumer-Verfügbarkeit. $10 / $50 pro 1M Tokens." }, "role": { "en": "Restricted Mythos-tier access for cyber defenders, critical infrastructure, and approved biology researchers — not a default deployment model.", "de": "Eingeschränkter Mythos-Tier-Zugang für Cyberdefender, kritische Infrastruktur und genehmigte Biologie-Forscher — kein Standard-Deployment-Modell." }, "releaseDate": "2026-06-09", "knowledgeCutoff": "2026-04-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 10, "outputPriceUSD": 50, "pricingNote": { "en": "Same list pricing as Fable 5. Access requires Anthropic trusted-access approval; 30-day traffic retention applies per the launch policy.", "de": "Gleicher Listenpreis wie Fable 5. Zugang erfordert Anthropic-Trusted-Access-Freigabe; 30-Tage-Traffic-Retention gilt laut Launch-Policy." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-fable-5-mythos-5", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/claude-mythos-5" }, { "slug": "claude-mythos-preview", "name": "Claude Mythos Preview", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's June 2026 research preview. Also leads GPQA Diamond (94.6%), HLE (64.7%), and CharXiv (93.2%). 2M context, $75 / $300 per 1M tokens.", "de": "Anthropics Research-Preview Juni 2026. Führt ausserdem GPQA Diamond (94,6 %), HLE (64,7 %) und CharXiv (93,2 %) an. 2M-Kontext, $75 / $300 pro 1M Tokens." }, "role": { "en": "Research preview for the hardest reasoning, coding, and generalization tasks. Not yet in general availability — access via the Claude research program. Treat as a research artifact, not a deployment candidate.", "de": "Research-Preview für die härtesten Reasoning-, Coding- und Generalisierungsaufgaben. Noch nicht allgemein verfügbar — Zugang über das Claude-Research-Programm. Als Research-Artefakt behandeln, nicht als Deployment-Kandidat." }, "releaseDate": "2026-06-01", "knowledgeCutoff": "2026-04-01", "contextWindow": 2000000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 75, "outputPriceUSD": 300, "pricingNote": { "en": "Research-preview pricing reflects a frontier model with high inference cost. Anthropic states this is a temporary tier until Mythos reaches general availability.", "de": "Research-Preview-Preise spiegeln ein Frontier-Modell mit hohen Inferenzkosten wider. Anthropic gibt an, dass dies ein temporärer Tarif ist, bis Mythos die allgemeine Verfügbarkeit erreicht." }, "pricingChecked": "2026-06-02", "status": "flagship", "sources": [ { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-02" }, { "label": { "en": "ARC Prize leaderboard", "de": "ARC-Prize-Leaderboard" }, "url": "https://arcprize.org/leaderboard", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/claude-mythos-preview" }, { "slug": "claude-opus-4-2024-07", "name": "Claude Opus 4 (2024-07)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "July 2024 Claude Opus 4 checkpoint. Predecessor to Opus 4.5+.", "de": "Claude-Opus-4-Checkpoint Juli 2024. Vorgänger von Opus 4.5+." }, "role": { "en": "Early Opus 4 API snapshot for leaderboard continuity.", "de": "Früher Opus-4-API-Snapshot für Leaderboard-Kontinuität." }, "releaseDate": "2024-07-01", "knowledgeCutoff": "2024-04-01", "contextWindow": 200000, "maxOutput": 32000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 15, "outputPriceUSD": 75, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Claude 4 family", "de": "Claude-4-Familie" }, "url": "https://www.anthropic.com/news/claude-4", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/claude-opus-4-2024-07" }, { "slug": "claude-opus-4-5-thinking", "name": "Claude Opus 4.5 (Reasoning)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-11-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/claude-opus-4-5-thinking" }, { "slug": "claude-opus-4-5", "name": "Claude Opus 4.5", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's November 2025 flagship. Best at coding and agents at its release. $5/$25 per 1M tokens, 200K context. Superseded by Opus 4.6/4.7.", "de": "Anthropics Flaggschiff November 2025. Bei Release das Beste für Coding und Agenten. $5/$25 pro 1M Tokens, 200K-Kontext. Durch Opus 4.6/4.7 abgelöst." }, "role": { "en": "Previous-gen Opus for coding and agentic tasks. Still available but Opus 4.7 is the current default.", "de": "Vorherige Opus-Generation für Coding und Agent-Aufgaben. Noch verfügbar, aber Opus 4.7 ist das aktuelle Standard." }, "releaseDate": "2025-11-24", "knowledgeCutoff": "2025-10-01", "contextWindow": 200000, "maxOutput": 64000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingNote": { "en": "Same price as Opus 4.6/4.7. 67% cheaper than the legacy Opus 4.1.", "de": "Gleicher Preis wie Opus 4.6/4.7. 67 % billiger als das alte Opus 4.1." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-opus-4-5", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/claude-opus-4-5" }, { "slug": "claude-opus-4-6-adaptive", "name": "Claude Opus 4.6 (Adaptive Reasoning, Max Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-05", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/claude-opus-4-6-adaptive" }, { "slug": "claude-opus-4-6", "name": "Claude Opus 4.6", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's February 2026 flagship. First Opus with 1M context (beta), 128K output, multi-agent teams in Claude Code. $5/$25 per 1M tokens.", "de": "Anthropics Flaggschiff Februar 2026. Erstes Opus mit 1M-Kontext (Beta), 128K-Output, Multi-Agent-Teams in Claude Code. $5/$25 pro 1M Tokens." }, "role": { "en": "Frontier Opus model for complex reasoning, multi-step coding, and massive context tasks. Superseded by Opus 4.7 but still available.", "de": "Frontier-Opus-Modell für komplexes Reasoning, mehrstufiges Coding und Aufgaben mit riesigem Kontext. Durch Opus 4.7 abgelöst, aber noch verfügbar." }, "releaseDate": "2026-02-05", "knowledgeCutoff": "2025-12-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingNote": { "en": "1M context at standard pricing — no long-context surcharge. Prompt caching saves up to 90%.", "de": "1M-Kontext zum Standardpreis — kein Aufschlag für langen Kontext. Prompt-Caching spart bis zu 90 %." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-opus-4-6", "checkedAt": "2026-06-02" }, { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/claude-opus-4-6" }, { "slug": "claude-opus-4-7-non-reasoning", "name": "Claude Opus 4.7 (Non-reasoning, High Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-16", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/claude-opus-4-7-non-reasoning" }, { "slug": "claude-opus-4-7", "name": "Claude Opus 4.7", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's April 2026 flagship. 13% coding lift over Opus 4.6, 3x more production tasks resolved, high-res vision up to 3.75 MP, new tokenizer. $5/$25 per 1M tokens, 1M context window.", "de": "Anthropics Flaggschiff April 2026. 13 % Coding-Steigerung gegenüber Opus 4.6, 3× mehr Produktionsaufgaben gelöst, hochauflösendes Vision bis 3,75 MP, neuer Tokenizer. $5/$25 pro 1M Tokens, 1M-Kontextfenster." }, "role": { "en": "Most capable generally available Claude model. Best for complex reasoning, long-running agentic tasks, and high-stakes code review.", "de": "Fähigstes allgemein verfügbares Claude-Modell. Ideal für komplexe Reasoning-, lang laufende Agent-Aufgaben und Code-Review mit hohem Einsatz." }, "releaseDate": "2026-04-16", "knowledgeCutoff": "2026-02-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingNote": { "en": "Same price as Opus 4.6. Prompt caching saves up to 90% on cached input. Batch API cuts costs 50%.", "de": "Gleicher Preis wie Opus 4.6. Prompt-Caching spart bis zu 90 % bei gecachtem Input. Batch-API reduziert Kosten um 50 %." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "Anthropic announcement", "de": "Anthropic-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-opus-4-7", "checkedAt": "2026-06-02" }, { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/claude-opus-4-7" }, { "slug": "claude-opus-4-8", "name": "Claude Opus 4.8", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's May 2026 flagship.", "de": "Anthropics Flaggschiff Mai 2026." }, "role": { "en": "Front-door model for hard reasoning, long documents, agentic coding, and tool-heavy research. The model we reach for first when a task is too hard for Sonnet but small enough to fit in one prompt.", "de": "Eingangsmodell für hartes Reasoning, lange Dokumente, agentisches Coding und tool-lastige Recherche. Das Modell, zu dem wir zuerst greifen, wenn eine Aufgabe zu schwer für Sonnet, aber klein genug für einen Prompt ist." }, "releaseDate": "2026-05-28", "knowledgeCutoff": "2026-02-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 15, "outputPriceUSD": 75, "pricingNote": { "en": "Prompt caching, batch, and 1-hour-cache options cut effective list price up to 90%.", "de": "Prompt-Caching, Batch und 1-Stunden-Cache senken den Listenpreis effektiv um bis zu 90 %." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "Anthropic pricing", "de": "Anthropic-Preise" }, "url": "https://www.anthropic.com/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "Claude 4 family announcement", "de": "Claude-4-Family-Ankündigung" }, "url": "https://www.anthropic.com/news/claude-4", "checkedAt": "2026-06-01" }, { "label": { "en": "Artificial Analysis", "de": "Artificial Analysis" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/claude-opus-4-8" }, { "slug": "claude-opus-5-5", "name": "Claude Opus 5.5", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's 22 Sep 2026 Opus. API list $4 / $20 per 1M tokens, 1M context, 128K output. Text and image in, text out. No AA or Vals ledger row on this page as of 23 Sep 2026.", "de": "Anthropics Opus vom 22. Sep. 2026. API-Liste $4 / $20 pro 1M Tokens, 1M-Kontext, 128K-Output. Text und Bild rein, Text raus. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite Stand 23. Sep. 2026." }, "role": { "en": "Current Claude Opus line for long-running coding and knowledge work on the Claude API, Claude apps, and major clouds.", "de": "Aktuelle Claude-Opus-Linie für langlaufendes Coding und Wissensarbeit über die Claude-API, die Claude-Apps und die grossen Clouds." }, "releaseDate": "2026-09-22", "knowledgeCutoff": "Jun 2026", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 4, "outputPriceUSD": 20, "pricingNote": { "en": "Claude Platform list checked 23 Sep 2026: $4 input / $20 output per 1M tokens. Cache reads $0.20. Five-minute cache writes $5; one-hour cache writes $8. Fast mode is $8 input / $40 output. Batch API is half of input and output. Thinking stays on. Reliable knowledge cutoff is June 2026.", "de": "Claude-Platform-Liste geprüft am 23. Sep. 2026: $4 Input / $20 Output pro 1M Tokens. Cache-Reads $0,20. Fünf-Minuten-Cache-Writes $5; Ein-Stunden-Cache-Writes $8. Fast-Mode ist $8 Input / $40 Output. Die Batch-API berechnet die Hälfte von Input und Output. Thinking bleibt an. Verlässlicher Knowledge-Cutoff ist Juni 2026." }, "pricingChecked": "2026-09-28", "status": "flagship", "sources": [ { "label": { "en": "Anthropic Claude Opus 5.5 announcement", "de": "Anthropic-Ankündigung Claude Opus 5.5" }, "url": "https://www.anthropic.com/claude-opus-5-5", "checkedAt": "2026-09-28" }, { "label": { "en": "Claude Platform Opus 5.5 overview", "de": "Claude-Platform-Übersicht Opus 5.5" }, "url": "https://platform.claude.com/docs/en/models/opus-5-5/overview", "checkedAt": "2026-09-28" }, { "label": { "en": "Claude Platform pricing", "de": "Claude-Platform-Preise" }, "url": "https://platform.claude.com/docs/en/about-claude/pricing", "checkedAt": "2026-09-28" } ], "html": "https://verdictpal.com/models/claude-opus-5-5" }, { "slug": "claude-opus-5-high", "name": "Claude Opus 5 (Adaptive Reasoning, High Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/claude-opus-5-high" }, { "slug": "claude-opus-5-low", "name": "Claude Opus 5 (Adaptive Reasoning, Low Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/claude-opus-5-low" }, { "slug": "claude-opus-5-medium", "name": "Claude Opus 5 (Adaptive Reasoning, Medium Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/claude-opus-5-medium" }, { "slug": "claude-opus-5-xhigh", "name": "Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/claude-opus-5-xhigh" }, { "slug": "claude-opus-5", "name": "Claude Opus 5 (Adaptive Reasoning, Max Effort)", "provider": "Anthropic", "family": "Claude Opus", "summary": { "en": "Anthropic's July 2026 Opus line. API list $5 / $25 per 1M tokens.", "de": "Anthropics Opus-Linie vom Juli 2026. API-Liste $5 / $25 pro 1M Tokens." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-24", "knowledgeCutoff": "open", "contextWindow": 1000000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingChecked": "2026-09-05", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models/claude-opus-5", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/claude-opus-5" }, { "slug": "claude-sonnet-4-6-adaptive", "name": "Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $3/$15 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-17", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/claude-sonnet-4-6-adaptive" }, { "slug": "claude-sonnet-4-6-non-reasoning-low-effort", "name": "Claude Sonnet 4.6 (Non-reasoning, Low Effort)", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $3/$15 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-17", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/claude-sonnet-4-6-non-reasoning-low-effort" }, { "slug": "claude-sonnet-4-6", "name": "Claude Sonnet 4.6", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "Anthropic's February 2026 mid-tier model.", "de": "Anthropics Mittelklassemodell Februar 2026." }, "role": { "en": "The default for production work that needs Anthropic's voice and tool-use discipline without the Opus 4.8 price tag.", "de": "Standard für Produktionsarbeit, die Anthropics Stimme und Tool-Use-Disziplin braucht, ohne den Opus-4.8-Preis." }, "releaseDate": "2026-02-17", "knowledgeCutoff": "2025-12-01", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingNote": { "en": "5× cheaper than Opus 4.8 on input and 5× on output. Prompt caching and batch discount further cut effective list price.", "de": "5× günstiger als Opus 4.8 beim Input und 5× beim Output. Prompt-Caching und Batch-Rabatt senken den effektiven Listenpreis weiter." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "Anthropic pricing", "de": "Anthropic-Preise" }, "url": "https://www.anthropic.com/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "Sonnet 4.6 release", "de": "Sonnet-4.6-Release" }, "url": "https://www.anthropic.com/news/claude-4-6", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/claude-sonnet-4-6" }, { "slug": "claude-sonnet-5-5", "name": "Claude Sonnet 5.5", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "Anthropic's 28 Sep 2026 Sonnet. API list $2 / $10 per 1M tokens, 1M context, 128K output. Text and image in, text out. Reliable knowledge cutoff June 2026. No AA or Vals ledger row on this page.", "de": "Anthropics Sonnet vom 28. Sep. 2026. API-Liste $2 / $10 pro 1M Tokens, 1M-Kontext, 128K-Output. Text und Bild rein, Text raus. Verlässlicher Knowledge-Cutoff Juni 2026. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite." }, "role": { "en": "Current Claude Sonnet line for everyday coding, documents, and fast iteration on the Claude API, Claude apps, and major clouds.", "de": "Aktuelle Claude-Sonnet-Linie für alltägliches Coding, Dokumente und schnelle Iteration über die Claude-API, die Claude-Apps und die grossen Clouds." }, "releaseDate": "2026-09-28", "knowledgeCutoff": "Jun 2026", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 10, "pricingNote": { "en": "Claude Platform list checked 28 Sep 2026: $2 input / $10 output per 1M tokens. Cache reads $0.20. Five-minute cache writes $2.50; one-hour cache writes $4. Batch API is half of input and output. Default API effort is high. Reliable knowledge cutoff is June 2026.", "de": "Claude-Platform-Liste geprüft am 28. Sep. 2026: $2 Input / $10 Output pro 1M Tokens. Cache-Reads $0,20. Fünf-Minuten-Cache-Writes $2,50; Ein-Stunden-Cache-Writes $4. Die Batch-API berechnet die Hälfte von Input und Output. Standard-Effort auf der API ist high. Verlässlicher Knowledge-Cutoff ist Juni 2026." }, "pricingChecked": "2026-09-28", "status": "flagship", "sources": [ { "label": { "en": "Anthropic Claude Sonnet 5.5 announcement", "de": "Anthropic-Ankündigung Claude Sonnet 5.5" }, "url": "https://www.anthropic.com/claude-sonnet-5-5", "checkedAt": "2026-09-28" }, { "label": { "en": "Claude Platform Sonnet 5.5 overview", "de": "Claude-Platform-Übersicht Sonnet 5.5" }, "url": "https://platform.claude.com/docs/en/models/sonnet-5-5/overview", "checkedAt": "2026-09-28" }, { "label": { "en": "Claude Platform pricing", "de": "Claude-Platform-Preise" }, "url": "https://platform.claude.com/docs/en/about-claude/pricing", "checkedAt": "2026-09-28" } ], "html": "https://verdictpal.com/models/claude-sonnet-5-5" }, { "slug": "claude-sonnet-5", "name": "Claude Sonnet 5", "provider": "Anthropic", "family": "Claude Sonnet", "summary": { "en": "Anthropic's June 2026 Sonnet-class release. API list price $2 / $10 per 1M tokens.", "de": "Anthropics Sonnet-Klasse Juni 2026. API-Listenpreis $2 / $10 pro 1M Tokens." }, "role": { "en": "Previous Sonnet line, still on the API. The 28 Sep 2026 release is the current mid-tier. Use max-effort AA rows for intelligence comparisons; lower-effort variants score differently on the same suite.", "de": "Vorherige Sonnet-Linie, weiter auf der API. Das Release vom 28. Sep. 2026 ist die aktuelle Mittelklasse. Nutze Max-Effort-AA-Zeilen für Intelligenzvergleiche; niedrigere Efforts scoren auf derselben Suite anders." }, "releaseDate": "2026-06-30", "knowledgeCutoff": "Jan 2026", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 10, "pricingNote": { "en": "AA list pricing for Claude Sonnet 5 (adaptive max effort) as of 2026-07-09: $2 input / $10 output per 1M tokens.", "de": "AA-Listenpreise für Claude Sonnet 5 (adaptiv, Max Effort) Stand 2026-07-09: $2 Input / $10 Output pro 1M Tokens." }, "pricingChecked": "2026-09-28", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" }, { "label": { "en": "Anthropic Claude", "de": "Anthropic Claude" }, "url": "https://www.anthropic.com/claude", "checkedAt": "2026-07-09" }, { "label": { "en": "Claude Platform Sonnet 5 overview", "de": "Claude-Platform-Übersicht Sonnet 5" }, "url": "https://platform.claude.com/docs/en/models/sonnet-5/overview", "checkedAt": "2026-09-28" }, { "label": { "en": "Claude Platform pricing", "de": "Claude-Platform-Preise" }, "url": "https://platform.claude.com/docs/en/about-claude/pricing", "checkedAt": "2026-09-28" } ], "html": "https://verdictpal.com/models/claude-sonnet-5" }, { "slug": "cohere-command-a", "name": "Command A", "provider": "Cohere", "family": "Command", "summary": { "en": "Cohere's enterprise-focused Command tier. Strong at RAG, tool use, and multilingual retrieval workflows. 256K context.", "de": "Cohere's Enterprise-fokussiertes Command-Tier. Stark bei RAG, Tool-Use und mehrsprachigen Retrieval-Workflows. 256K-Kontext." }, "role": { "en": "Pick when your stack already uses Cohere embed + rerank and you want one vendor for retrieval pipelines.", "de": "Wahl, wenn der Stack bereits Cohere Embed + Rerank nutzt und ein Anbieter für Retrieval-Pipelines reicht." }, "releaseDate": "2026-03-15", "knowledgeCutoff": "2025-06-01", "contextWindow": 256000, "maxOutput": 8192, "modalities": [ "text", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 2.5, "outputPriceUSD": 10, "pricingNote": { "en": "Enterprise tier — confirm current rates on Cohere dashboard; volume discounts common.", "de": "Enterprise-Tier — aktuelle Tarife im Cohere-Dashboard prüfen; Volumenrabatte üblich." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Cohere models", "de": "Cohere-Modelle" }, "url": "https://docs.cohere.com/docs/models", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/cohere-command-a" }, { "slug": "composer-2-5-fast", "name": "Composer 2.5 Fast", "provider": "Cursor", "family": "Composer", "summary": { "en": "Cursor's default IDE agent model (May 2026).", "de": "Cursors Default-IDE-Agent-Modell (Mai 2026)." }, "role": { "en": "Default Composer tier in Cursor for live pair-programming, multi-file agent applies, and terminal tool loops where responsiveness beats batch cost.", "de": "Default-Composer-Tier in Cursor für Live-Pair-Programming, Multi-File-Agent-Applies und Terminal-Tool-Loops, wo Reaktionszeit Batch-Kosten schlägt." }, "releaseDate": "2026-05-18", "knowledgeCutoff": "2026-03-01", "contextWindow": 200000, "maxOutput": 64000, "modalities": [ "text", "code", "tool-use" ], "access": [ "consumer-app", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingNote": { "en": "Fast is the IDE default. Standard tier is $0.50 / $2.50 per 1M tokens for background or CI workloads.", "de": "Fast ist der IDE-Default. Standard-Tier: $0,50 / $2,50 pro 1M Tokens für Background- oder CI-Workloads." }, "pricingChecked": "2026-06-10", "status": "flagship", "sources": [ { "label": { "en": "Cursor Composer 2.5 docs", "de": "Cursor Composer-2.5-Docs" }, "url": "https://cursor.com/docs/models/cursor-composer-2-5", "checkedAt": "2026-06-10" }, { "label": { "en": "AA Composer 2.5 analysis", "de": "AA Composer-2.5-Analyse" }, "url": "https://artificialanalysis.ai/articles/cursor-composer-2-5-coding-agent-index", "checkedAt": "2026-06-10" } ], "html": "https://verdictpal.com/models/composer-2-5-fast" }, { "slug": "composer-2-5", "name": "Composer 2.5 Standard", "provider": "Cursor", "family": "Composer", "summary": { "en": "Cursor's cost-optimized Composer tier.", "de": "Cursors kostenoptimiertes Composer-Tier." }, "role": { "en": "Batch and CI Composer tier when you want the same agentic coding intelligence without paying the Fast inference premium.", "de": "Batch- und CI-Composer-Tier, wenn dieselbe agentische Coding-Intelligenz ohne Fast-Inference-Aufpreis reichen soll." }, "releaseDate": "2026-05-18", "knowledgeCutoff": "2026-03-01", "contextWindow": 200000, "maxOutput": 64000, "modalities": [ "text", "code", "tool-use" ], "access": [ "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.5, "outputPriceUSD": 2.5, "pricingNote": { "en": "Select Standard in the Composer model picker; Fast remains the product default.", "de": "Standard im Composer-Modell-Picker wählen; Fast bleibt der Produkt-Default." }, "pricingChecked": "2026-06-10", "status": "solid", "sources": [ { "label": { "en": "Cursor Composer 2.5 docs", "de": "Cursor Composer-2.5-Docs" }, "url": "https://cursor.com/docs/models/cursor-composer-2-5", "checkedAt": "2026-06-10" }, { "label": { "en": "AA Composer 2.5 analysis", "de": "AA Composer-2.5-Analyse" }, "url": "https://artificialanalysis.ai/articles/cursor-composer-2-5-coding-agent-index", "checkedAt": "2026-06-10" } ], "html": "https://verdictpal.com/models/composer-2-5" }, { "slug": "deepseek-coder-v2", "name": "DeepSeek Coder V2", "provider": "DeepSeek", "family": "DeepSeek Coder", "summary": { "en": "DeepSeek's May 2024 coding model. Baseline before V4 and R1.", "de": "DeepSeeks Coding-Modell Mai 2024. Baseline vor V4 und R1." }, "role": { "en": "Legacy open-weights coding route — prefer DeepSeek V4 or R1.", "de": "Legacy-Open-Weights-Coding-Route — DeepSeek V4 oder R1 bevorzugen." }, "releaseDate": "2024-05-15", "knowledgeCutoff": "open", "contextWindow": 128000, "maxOutput": 8192, "modalities": [ "text", "code" ], "access": [ "open-weights", "api", "managed-cloud" ], "inputPriceUSD": 0.14, "outputPriceUSD": 0.28, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "DeepSeek Coder V2", "de": "DeepSeek Coder V2" }, "url": "https://github.com/deepseek-ai/DeepSeek-Coder-V2", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/deepseek-coder-v2" }, { "slug": "deepseek-r1", "name": "DeepSeek R1", "provider": "DeepSeek", "family": "DeepSeek R1", "summary": { "en": "DeepSeek's January 2025 reasoning model. Open weights, chain-of-thought style.", "de": "DeepSeeks Reasoning-Modell Januar 2025. Open Weights, Chain-of-Thought-Stil." }, "role": { "en": "Self-hosted reasoning route before V4 shipped. Still relevant for reproducible CoT research and local inference stacks.", "de": "Self-Hosting-Reasoning-Route vor V4. Weiter relevant für reproduzierbare CoT-Forschung und lokale Inference-Stacks." }, "releaseDate": "2025-01-20", "knowledgeCutoff": "open", "contextWindow": 128000, "maxOutput": 32768, "modalities": [ "text", "code", "reasoning" ], "access": [ "open-weights", "api", "managed-cloud" ], "inputPriceUSD": 0.55, "outputPriceUSD": 2.19, "pricingNote": { "en": "Legacy API pricing; V4 Flash supersedes for new production work.", "de": "Legacy-API-Preise; V4 Flash löst für neue Produktionsarbeit ab." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "DeepSeek R1 model card", "de": "DeepSeek-R1-Model-Karte" }, "url": "https://huggingface.co/deepseek-ai/DeepSeek-R1", "checkedAt": "2026-06-09" }, { "label": { "en": "DeepSeek R1 paper", "de": "DeepSeek-R1-Paper" }, "url": "https://arxiv.org/abs/2501.12948", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/deepseek-r1" }, { "slug": "deepseek-v3-2-reasoning", "name": "DeepSeek V3.2 (Reasoning)", "provider": "DeepSeek", "family": "DeepSeek V3.2", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.28/$0.42 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.28/$0.42 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-01", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.28, "outputPriceUSD": 0.42, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/deepseek-v3-2-reasoning" }, { "slug": "deepseek-v4-1-flash", "name": "DeepSeek V4.1 Flash", "provider": "DeepSeek", "family": "DeepSeek V4.1", "summary": { "en": "DeepSeek's 10 Sep 2026 Flash refresh: 552B MoE, 8B active on prefill and 16B on decode, native vision, 1M context, 384K max output. API id deepseek-flash. Official off-peak list is $0.15 / $0.60 per 1M tokens (cache-miss input / output); peak doubles. Retired V4-Flash aliases route here.", "de": "DeepSeeks Flash-Refresh vom 10. Sep. 2026: 552B-MoE, 8B aktiv beim Prefill und 16B beim Decode, native Vision, 1M-Kontext, 384K max. Output. API-ID deepseek-flash. Offizielle Off-Peak-Liste $0,15 / $0,60 pro 1M Tokens (Cache-Miss-Input / Output); Peak verdoppelt. Ausgemusterte V4-Flash-Aliasse landen hier." }, "role": { "en": "Current DeepSeek Flash API model for cheap multimodal and agent work. Prefer official pricing and architecture notes over vendor eval tables; AA rows land when the snapshot lists the slug.", "de": "Aktuelles DeepSeek-Flash-API-Modell für günstige multimodale und agentische Arbeit. Bevorzuge offizielle Preise und Architektur-Notizen gegenüber Vendor-Eval-Tabellen; AA-Zeilen kommen, wenn der Snapshot den Slug listet." }, "releaseDate": "2026-09-10", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "maxOutput": 384000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "open-weights" ], "inputPriceUSD": 0.15, "outputPriceUSD": 0.6, "pricingNote": { "en": "Official DeepSeek list from 10 Sep 2026, per 1M tokens. Off-peak: cache-hit input $0.003, cache-miss input $0.15, output $0.60. Peak (weekday UTC windows) doubles each meter. Call deepseek-flash. Legacy deepseek-v4-flash and deepseek-v4-flash-vision-exp route here at Flash rates. The Sep 10 plan to route deepseek-v4-pro here on 14 Sep was withdrawn; Pro stays on its own API and rates.", "de": "Offizielle DeepSeek-Liste ab 10. Sep. 2026, pro 1M Tokens. Off-Peak: Cache-Hit-Input $0,003, Cache-Miss-Input $0,15, Output $0,60. Peak (Wochentags-UTC-Fenster) verdoppelt jeden Zähler. Aufruf als deepseek-flash. Legacy deepseek-v4-flash und deepseek-v4-flash-vision-exp landen hier zu Flash-Preisen. Der Plan vom 10. Sep., deepseek-v4-pro am 14. Sep. hierher zu routen, wurde zurückgezogen; Pro bleibt auf eigener API und eigenen Tarifen." }, "pricingChecked": "2026-09-18", "status": "solid", "sources": [ { "label": { "en": "DeepSeek V4.1 Flash announcement", "de": "DeepSeek-V4.1-Flash-Ankündigung" }, "url": "https://www.deepseek.com/en/news/deepseek-v4-1-flash/", "checkedAt": "2026-09-18" }, { "label": { "en": "DeepSeek API models and pricing", "de": "DeepSeek-API-Modelle und Preise" }, "url": "https://api-docs.deepseek.com/quick_start/pricing", "checkedAt": "2026-09-18" }, { "label": { "en": "DeepSeek API changelog 10 Sep 2026", "de": "DeepSeek-API-Changelog 10. Sep. 2026" }, "url": "https://api-docs.deepseek.com/news/news260910", "checkedAt": "2026-09-18" }, { "label": { "en": "Hugging Face DeepSeek-V4.1-Flash", "de": "Hugging-Face-DeepSeek-V4.1-Flash" }, "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash", "checkedAt": "2026-09-12" } ], "html": "https://verdictpal.com/models/deepseek-v4-1-flash" }, { "slug": "deepseek-v4-flash-high", "name": "DeepSeek V4 Flash (Reasoning, High Effort)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.14/$0.28 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.14/$0.28 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.14, "outputPriceUSD": 0.28, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/deepseek-v4-flash-high" }, { "slug": "deepseek-v4-flash-max", "name": "DeepSeek V4 Flash Max", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "DeepSeek's April 2026 cost-efficient V4 tier. 1M context, $0.14/$0.28 per 1M tokens (cache miss).", "de": "DeepSeeks kosteneffizientes V4-Tier April 2026. 1M-Kontext, $0,14/$0,28 pro 1M Tokens (Cache Miss)." }, "role": { "en": "Cheapest long-context coding route on the public ledger. Default DeepSeek API pick for agents.", "de": "Günstigste Long-Context-Coding-Route im öffentlichen Ledger. Standard-DeepSeek-API-Wahl für Agenten." }, "releaseDate": "2026-04-24", "knowledgeCutoff": "open", "contextWindow": 1000000, "maxOutput": 384000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.14, "outputPriceUSD": 0.28, "pricingNote": { "en": "Cache-hit input $0.0028/1M. Open weights under MIT license.", "de": "Cache-Hit-Input $0,0028/1M. Open Weights unter MIT-Lizenz." }, "pricingChecked": "2026-06-09", "status": "flagship", "sources": [ { "label": { "en": "DeepSeek API pricing", "de": "DeepSeek-API-Preise" }, "url": "https://api-docs.deepseek.com/quick_start/pricing", "checkedAt": "2026-06-09" }, { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/deepseek-v4-flash-max" }, { "slug": "deepseek-v4-flash-non-reasoning", "name": "DeepSeek V4 Flash (Non-reasoning)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.14/$0.28 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.14/$0.28 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-24", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.14, "outputPriceUSD": 0.28, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/deepseek-v4-flash-non-reasoning" }, { "slug": "deepseek-v4-flash-vision", "name": "DeepSeek V4 Flash Vision (Reasoning, Max Effort)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.44/$1.32 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.44/$1.32 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-21", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "image", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.44, "outputPriceUSD": 1.32, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/deepseek-v4-flash-vision" }, { "slug": "deepseek-v4-flash", "name": "DeepSeek V4 Flash 0731 (Reasoning, Max Effort)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.44/$1.32 per 1M tokens after the 17 Aug 2026 list-price change. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0,44/$1,32 pro 1M Tokens nach der Listenpreis-Änderung vom 17. Aug. 2026. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-31", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.44, "outputPriceUSD": 1.32, "pricingNote": { "en": "AA list price for the 0731 Max Effort snapshot after DeepSeek's 17 Aug 2026 V4 API change. Older 0420 Flash snapshots stay on their own model pages.", "de": "AA-Listenpreis für den 0731-Max-Effort-Snapshot nach DeepSeeks V4-API-Änderung vom 17. Aug. 2026. Ältere 0420-Flash-Snapshots bleiben auf eigenen Seiten." }, "pricingChecked": "2026-08-17", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis DeepSeek V4 Flash", "de": "Artificial Analysis DeepSeek V4 Flash" }, "url": "https://artificialanalysis.ai/models/deepseek-v4-flash", "checkedAt": "2026-08-17" } ], "html": "https://verdictpal.com/models/deepseek-v4-flash" }, { "slug": "deepseek-v4-pro-high", "name": "DeepSeek V4 Pro (Reasoning, High Effort)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.435/$0.87 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.435/$0.87 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-24", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.435, "outputPriceUSD": 0.87, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/deepseek-v4-pro-high" }, { "slug": "deepseek-v4-pro-max", "name": "DeepSeek V4 Pro Max", "provider": "DeepSeek", "family": "DeepSeek", "summary": { "en": "Cheapest frontier pricing on the public market.", "de": "Günstigster Frontier-Preis auf dem öffentlichen Markt." }, "role": { "en": "Default model for self-hosted agent infrastructure, cost-sensitive deployments, and teams that need open weights with frontier-tier coding scores.", "de": "Default-Modell für selbstgehostete Agent-Infrastruktur, kostenkritische Deployments und Teams, die Open Weights mit Frontier-Coding-Scores brauchen." }, "releaseDate": "2026-04-24", "knowledgeCutoff": "2026-02-01", "contextWindow": 256000, "maxOutput": 65536, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.4, "outputPriceUSD": 1.6, "pricingNote": { "en": "API and self-host both available. Open weights are released under a permissive license for self-hosting at scale.", "de": "API und Self-Host beide verfügbar. Open Weights unter permissiver Lizenz für Self-Hosting im Massstab veröffentlicht." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "DeepSeek API pricing", "de": "DeepSeek-API-Preise" }, "url": "https://api-docs.deepseek.com/quick_start/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "V4 Pro Max model card", "de": "V4-Pro-Max-Model-Karte" }, "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro-Max", "checkedAt": "2026-06-01" }, { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/deepseek-v4-pro-max" }, { "slug": "deepseek-v4-pro-non-reasoning", "name": "DeepSeek V4 Pro (Non-reasoning)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.435/$0.87 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.435/$0.87 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-24", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.435, "outputPriceUSD": 0.87, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/deepseek-v4-pro-non-reasoning" }, { "slug": "deepseek-v4-pro", "name": "DeepSeek V4 Pro 0813 (Reasoning, Max Effort)", "provider": "DeepSeek", "family": "DeepSeek V4", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.32/$3.96 per 1M tokens after the 17 Aug 2026 list-price change. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1,32/$3,96 pro 1M Tokens nach der Listenpreis-Änderung vom 17. Aug. 2026. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-13", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.32, "outputPriceUSD": 3.96, "pricingNote": { "en": "AA hosted peak list for the 0813 Max Effort snapshot is $1.32 / $3.96 per 1M tokens. Official DeepSeek API (18 Sep 2026): off-peak cache-miss $0.66 / output $1.98; peak doubles those meters. The Sep 10 plan to retire deepseek-v4-pro onto V4.1 Flash on 14 Sep was withdrawn — Pro stays a separate API with unchanged billing until further notice. Older 0424 Pro snapshots stay on their own model pages.", "de": "AA-gehostete Peak-Liste für den 0813-Max-Effort-Snapshot ist $1,32 / $3,96 pro 1M Tokens. Offizielle DeepSeek-API (18. Sep. 2026): Off-Peak Cache-Miss $0,66 / Output $1,98; Peak verdoppelt diese Zähler. Der Plan vom 10. Sep., deepseek-v4-pro am 14. Sep. auf V4.1 Flash zu legen, wurde zurückgezogen — Pro bleibt eine separate API mit unveränderter Abrechnung bis auf Weiteres. Ältere 0424-Pro-Snapshots bleiben auf eigenen Seiten." }, "pricingChecked": "2026-09-18", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis DeepSeek V4 Pro", "de": "Artificial Analysis DeepSeek V4 Pro" }, "url": "https://artificialanalysis.ai/models/deepseek-v4-pro", "checkedAt": "2026-09-18" }, { "label": { "en": "DeepSeek API models and pricing", "de": "DeepSeek-API-Modelle und Preise" }, "url": "https://api-docs.deepseek.com/quick_start/pricing", "checkedAt": "2026-09-18" } ], "html": "https://verdictpal.com/models/deepseek-v4-pro" }, { "slug": "diffusiongemma-26b-a4b", "name": "DiffusionGemma 26B A4B", "provider": "Google", "family": "DiffusionGemma 26B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-06-10", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/diffusiongemma-26b-a4b" }, { "slug": "exaone-4-5-33b", "name": "EXAONE 4.5 33B", "provider": "Other", "family": "EXAONE 4.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-09", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/exaone-4-5-33b" }, { "slug": "g9v3-39a5b", "name": "G9v3-39A5B", "provider": "Other", "family": "G9v3", "summary": { "en": "AI9Stars G9v3-39A5B. AA lists $0 / $0 per 1M tokens.", "de": "AI9Stars G9v3-39A5B. AA führt $0 / $0 pro 1M Tokens." }, "role": { "en": "Open G9v3 MoE tracked on Artificial Analysis. Prefer sourced ledger rows; editorial depth is desk-pending.", "de": "Offenes G9v3-MoE auf Artificial Analysis. Bevorzuge belegte Ledger-Zeilen; redaktionelle Tiefe ist desk-pending." }, "releaseDate": "2026-08-03", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-15", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis G9v3-39A5B", "de": "Artificial-Analysis-G9v3-39A5B" }, "url": "https://artificialanalysis.ai/models/g9v3-39a5b", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/g9v3-39a5b" }, { "slug": "g9v3-3b", "name": "G9v3-3B", "provider": "Other", "family": "Other", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-23", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/g9v3-3b" }, { "slug": "gemini-1-0-ultra", "name": "Gemini 1.0 Ultra", "provider": "Google", "family": "Gemini 1", "summary": { "en": "Google's first Gemini Ultra tier. MMMU 59.4% on the public ledger. Historical multimodal baseline before Gemini 1.5.", "de": "Googles erstes Gemini-Ultra-Tier. MMMU 59,4 % im öffentlichen Ledger. Historische Multimodal-Baseline vor Gemini 1.5." }, "role": { "en": "Archive-only checkpoint for MMMU and early multimodal leaderboard rows.", "de": "Nur-Archiv-Checkpoint für MMMU und frühe Multimodal-Leaderboard-Zeilen." }, "releaseDate": "2024-02-08", "knowledgeCutoff": "2023-11-01", "contextWindow": 32768, "maxOutput": 8192, "modalities": [ "text", "image", "code" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Retired Ultra tier — not available on current Gemini API routes.", "de": "Ausgemustertes Ultra-Tier — nicht auf aktuellen Gemini-API-Routen verfügbar." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Gemini technical report", "de": "Gemini-Technical-Report" }, "url": "https://arxiv.org/abs/2312.11805", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-1-0-ultra" }, { "slug": "gemini-1-5-pro-2024-06", "name": "Gemini 1.5 Pro (2024-06)", "provider": "Google", "family": "Gemini 1.5", "summary": { "en": "June 2024 Gemini 1.5 Pro checkpoint. Long-context predecessor to Gemini 2/3.", "de": "Gemini-1.5-Pro-Checkpoint Juni 2024. Langkontext-Vorgänger von Gemini 2/3." }, "role": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger." }, "releaseDate": "2024-06-01", "knowledgeCutoff": "2024-04-01", "contextWindow": 2000000, "maxOutput": 8192, "modalities": [ "text", "image", "audio", "video", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Legacy checkpoint — confirm availability on Google AI Studio.", "de": "Legacy-Checkpoint — Verfügbarkeit in Google AI Studio prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Gemini 1.5 Pro", "de": "Gemini 1.5 Pro" }, "url": "https://ai.google.dev/gemini-api/docs/models/gemini", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-1-5-pro-2024-06" }, { "slug": "gemini-1-5-ultra-2024", "name": "Gemini 1.5 Ultra (2024)", "provider": "Google", "family": "Gemini 1.5", "summary": { "en": "2024 Gemini 1.5 Ultra. Pre-Gemini-2 flagship.", "de": "Gemini 1.5 Ultra 2024. Flaggschiff vor Gemini 2." }, "role": { "en": "Legacy Ultra tier for multimodal and knowledge benchmarks — superseded by Gemini 3.x.", "de": "Legacy-Ultra-Tier für Multimodal- und Wissens-Benchmarks — von Gemini 3.x abgelöst." }, "releaseDate": "2024-02-15", "knowledgeCutoff": "2023-11-01", "contextWindow": 1000000, "maxOutput": 8192, "modalities": [ "text", "image", "audio", "video", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Preview-era Ultra — check Google for archival access.", "de": "Preview-Ära-Ultra — Google für Archiv-Zugang prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Gemini technical report", "de": "Gemini-Technical-Report" }, "url": "https://arxiv.org/abs/2312.11805", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-1-5-ultra-2024" }, { "slug": "gemini-2-0-ultra", "name": "Gemini 2.0 Ultra", "provider": "Google", "family": "Gemini 2", "summary": { "en": "Google's late-2024 Ultra tier. Superseded by Gemini 3.x.", "de": "Googles Ultra-Tier Ende 2024. Von Gemini 3.x abgelöst." }, "role": { "en": "Legacy Gemini Ultra for long-context leaderboard comparisons.", "de": "Legacy-Gemini-Ultra für Langkontext-Leaderboard-Vergleiche." }, "releaseDate": "2024-12-11", "knowledgeCutoff": "2024-06-01", "contextWindow": 1000000, "maxOutput": 8192, "modalities": [ "text", "image", "audio", "video", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Preview-era pricing — check Google AI pricing for archival API access.", "de": "Preview-Ära-Preise — Google-AI-Preise für Archiv-API-Zugang prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Gemini 2.0 update", "de": "Gemini-2.0-Update" }, "url": "https://blog.google/technology/google-deepmind/google-gemini-ai-update-december-2024/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-2-0-ultra" }, { "slug": "gemini-3-1-flash-lite", "name": "Gemini 3.1 Flash Lite", "provider": "Google", "family": "Gemini 3.1", "summary": { "en": "Google's cheapest 1M-context Gemini. $0.25/$1.50 per 1M tokens. Volume tier for classification and routing.", "de": "Googles günstigstes 1M-Kontext-Gemini. $0,25/$1,50 pro 1M Tokens. Volumen-Tier für Klassifikation und Routing." }, "role": { "en": "High-volume preprocessing, tagging, and cheap RAG front doors before escalating to Pro.", "de": "Hochvolumiges Preprocessing, Tagging und günstige RAG-Eingänge vor Eskalation zu Pro." }, "releaseDate": "2026-05-07", "knowledgeCutoff": "2025-01-31", "contextWindow": 1048576, "maxOutput": 65536, "modalities": [ "text", "image", "video", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.25, "outputPriceUSD": 1.5, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Gemini API pricing", "de": "Gemini-API-Preise" }, "url": "https://ai.google.dev/pricing", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-3-1-flash-lite" }, { "slug": "gemini-3-1-pro-preview", "name": "Gemini 3.1 Pro Preview", "provider": "Google", "family": "Gemini 3.1", "summary": { "en": "Listed on Artificial Analysis with API pricing $2/$12 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $2/$12 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-19", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 2, "outputPriceUSD": 12, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/gemini-3-1-pro-preview" }, { "slug": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "provider": "Google", "family": "Gemini", "summary": { "en": "Google's February 2026 frontier model.", "de": "Googles Frontier-Modell Februar 2026." }, "role": { "en": "Google's strongest non-flash frontier model. The model to choose when 1M context, multimodal input, and tool-use all matter in the same workflow.", "de": "Googles stärkstes Frontier-Modell ohne Flash. Das Modell der Wahl, wenn 1M-Kontext, multimodaler Input und Tool-Use im selben Workflow zählen." }, "releaseDate": "2026-02-19", "knowledgeCutoff": "2026-01-31", "contextWindow": 1000000, "maxOutput": 65536, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 7, "outputPriceUSD": 21, "pricingNote": { "en": "Mid-tier pricing for a frontier model. 1M context comes standard. Native multimodal input for image, audio, and video.", "de": "Mittelklasse-Preis für ein Frontier-Modell. 1M-Kontext serienmässig. Nativer multimodaler Input für Bild, Audio und Video." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "Gemini API pricing", "de": "Gemini-API-Preise" }, "url": "https://ai.google.dev/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "Gemini 3.1 family", "de": "Gemini-3.1-Familie" }, "url": "https://ai.google.dev/gemini-3-1", "checkedAt": "2026-06-01" }, { "label": { "en": "Artificial Analysis", "de": "Artificial Analysis" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/gemini-3-1-pro" }, { "slug": "gemini-3-5-flash-lite", "name": "Gemini 3.5 Flash-Lite", "provider": "Google", "family": "Gemini 3.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-21", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 2.5, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/gemini-3-5-flash-lite" }, { "slug": "gemini-3-5-flash-medium", "name": "Gemini 3.5 Flash (medium)", "provider": "Google", "family": "Gemini 3.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.5/$9 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.5/$9 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-19", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.5, "outputPriceUSD": 9, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/gemini-3-5-flash-medium" }, { "slug": "gemini-3-5-flash-minimal", "name": "Gemini 3.5 Flash (minimal)", "provider": "Google", "family": "Gemini 3.5", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.5/$9 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.5/$9 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-19", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.5, "outputPriceUSD": 9, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/gemini-3-5-flash-minimal" }, { "slug": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "provider": "Google", "family": "Gemini", "summary": { "en": "Google's fastest frontier-tier model.", "de": "Googles schnellstes Frontier-Modell." }, "role": { "en": "Default model for high-throughput, low-latency tool use: search agents, batch classification, real-time customer-facing apps. The cost-per-1M gap to Opus 4.8 is roughly 50× on input and 60× on output.", "de": "Default-Modell für hohen Durchsatz und niedrige Latenz: Such-Agents, Batch-Klassifikation, Echtzeit-Kunden-Apps. Die Kosten-pro-1M-Lücke zu Opus 4.8 beträgt etwa 50× beim Input und 60× beim Output." }, "releaseDate": "2026-05-19", "knowledgeCutoff": "2026-03-01", "contextWindow": 1000000, "maxOutput": 65536, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingNote": { "en": "List pricing is the cheapest among May-2026 frontier models. Free tier available in AI Studio with rate limits.", "de": "Listenpreis ist der günstigste unter den Frontier-Modellen Mai 2026. Free-Tier in AI Studio mit Rate-Limits verfügbar." }, "pricingChecked": "2026-06-01", "status": "solid", "sources": [ { "label": { "en": "Gemini API pricing", "de": "Gemini-API-Preise" }, "url": "https://ai.google.dev/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "Gemini 3.5 family", "de": "Gemini-3.5-Familie" }, "url": "https://ai.google.dev/gemini-3", "checkedAt": "2026-06-01" }, { "label": { "en": "MCP Atlas leaderboard", "de": "MCP-Atlas-Leaderboard" }, "url": "https://github.com/mcp-atlas/leaderboard", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/gemini-3-5-flash" }, { "slug": "gemini-3-6-flash", "name": "Gemini 3.6 Flash (high)", "provider": "Google", "family": "Gemini 3.6", "summary": { "en": "Google introductory API price $0.75 / $3.75 per 1M tokens through 31 Dec 2026; AA still lists $1.50 / $7.50 on the 2026-08-13 snapshot.", "de": "Googles Einführungspreis $0,75 / $3,75 pro 1M Tokens bis 31. Dez. 2026; AA führt im Snapshot vom 2026-08-13 weiterhin $1,50 / $7,50." }, "role": { "en": "Prior Google Flash workhorse, superseded for most new agent work by 3.7 Flash. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Vorheriges Google-Flash-Arbeitspferd, für die meisten neuen Agent-Läufe von 3.7 Flash abgelöst. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-21", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "maxOutput": 64000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.75, "outputPriceUSD": 3.75, "pricingNote": { "en": "Google introductory list price for 3.6 Flash through 31 Dec 2026: $0.75 / $3.75 per 1M tokens, then $1.50 / $7.50. Artificial Analysis still listed $1.50 / $7.50 on the 2026-08-13 snapshot.", "de": "Googles Einführungs-Listenpreis für 3.6 Flash bis 31. Dez. 2026: $0,75 / $3,75 pro 1M Tokens, danach $1,50 / $7,50. Artificial Analysis führte im Snapshot vom 2026-08-13 weiterhin $1,50 / $7,50." }, "pricingChecked": "2026-08-13", "status": "solid", "sources": [ { "label": { "en": "DeepMind Gemini 3.6 Flash model card", "de": "DeepMind-Modellkarte Gemini 3.6 Flash" }, "url": "https://deepmind.google/models/model-cards/gemini-3-6-flash/", "checkedAt": "2026-08-13" }, { "label": { "en": "Artificial Analysis Gemini 3.6 Flash", "de": "Artificial-Analysis-Gemini-3.6-Flash" }, "url": "https://artificialanalysis.ai/models/gemini-3-6-flash", "checkedAt": "2026-08-13" }, { "label": { "en": "Google Cloud Gemini pricing", "de": "Google-Cloud-Gemini-Preise" }, "url": "https://cloud.google.com/gemini-enterprise-agent-platform/generative-ai/pricing", "checkedAt": "2026-08-13" } ], "html": "https://verdictpal.com/models/gemini-3-6-flash" }, { "slug": "gemini-3-7-flash", "name": "Gemini 3.7 Flash", "provider": "Google", "family": "Gemini 3.7", "summary": { "en": "Google's August 2026 Flash workhorse, superseded as the current Flash by Gemini 3.8 Flash on 2 Sep 2026. Introductory API price still $0.75 / $3.75 per 1M tokens through 31 Dec 2026.", "de": "Googles Flash-Arbeitspferd vom August 2026, als aktuelles Flash am 2. Sep. 2026 von Gemini 3.8 Flash abgelöst. Einführungspreis weiter $0,75 / $3,75 pro 1M Tokens bis 31. Dez. 2026." }, "role": { "en": "Previous Google Flash when 3.8 Flash is too verbose or you want the faster 305 t/s high-effort endpoint. Prefer sourced ledger rows; live AA v4.3 scores are not comparable to the August v4.1 snapshot.", "de": "Vorheriges Google-Flash, wenn 3.8 Flash zu wortreich ist oder der schnellere 305-t/s-High-Effort-Endpoint reicht. Bevorzuge belegte Ledger-Zeilen; live AA-v4.3-Scores sind nicht mit dem August-v4.1-Snapshot vergleichbar." }, "releaseDate": "2026-08-13", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "maxOutput": 64000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.75, "outputPriceUSD": 3.75, "pricingNote": { "en": "Google introductory list price through 31 Dec 2026: $0.75 input / $3.75 output per 1M tokens. From 1 Jan 2027 the stated list is $1.50 / $7.50. AA lists the same intro pair on the 2026-08-13 snapshot, with a 90% cache-hit discount on that page.", "de": "Googles Einführungs-Listenpreis bis 31. Dez. 2026: $0,75 Input / $3,75 Output pro 1M Tokens. Ab 1. Jan. 2027 gilt der angegebene Listenpreis $1,50 / $7,50. AA führt dasselbe Intro-Paar im Snapshot vom 2026-08-13, mit 90 % Cache-Hit-Rabatt auf der Modellseite." }, "pricingChecked": "2026-09-05", "status": "flagship", "sources": [ { "label": { "en": "Google Gemini 3.7 Flash announcement", "de": "Google-Gemini-3.7-Flash-Ankündigung" }, "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/", "checkedAt": "2026-08-13" }, { "label": { "en": "DeepMind Gemini 3.7 Flash model card", "de": "DeepMind-Modellkarte Gemini 3.7 Flash" }, "url": "https://deepmind.google/models/model-cards/gemini-3-7-flash/", "checkedAt": "2026-08-13" }, { "label": { "en": "Artificial Analysis Gemini 3.7 Flash", "de": "Artificial-Analysis-Gemini-3.7-Flash" }, "url": "https://artificialanalysis.ai/models/gemini-3-7-flash", "checkedAt": "2026-08-13" }, { "label": { "en": "Gemini API release notes", "de": "Gemini-API-Release-Notes" }, "url": "https://ai.google.dev/gemini-api/docs/changelog", "checkedAt": "2026-09-05" } ], "html": "https://verdictpal.com/models/gemini-3-7-flash" }, { "slug": "gemini-3-8-flash-cyber", "name": "Gemini 3.8 Flash Cyber", "provider": "Google", "family": "Gemini 3.8", "summary": { "en": "Restricted cybersecurity twin of Gemini 3.8 Flash. Same foundational intelligence; more permissive cyber mitigations. Fairwind Program only — not general API or consumer availability. Google cites CWE-Bench pass@1 47.2% (Collinear) as a vendor figure.", "de": "Eingeschränktes Cybersecurity-Zwilling von Gemini 3.8 Flash. Dieselbe Basis-Intelligenz; permissivere Cyber-Mitigations. Nur Fairwind-Programm — keine allgemeine API- oder Consumer-Verfügbarkeit. Google nennt CWE-Bench pass@1 47,2 % (Collinear) als Vendor-Zahl." }, "role": { "en": "Trusted-defender cyber model for vulnerability discovery and patching — not a default deployment model.", "de": "Trusted-Defender-Cyber-Modell für Schwachstellenfindung und Patching — kein Standard-Deployment-Modell." }, "releaseDate": "2026-09-02", "knowledgeCutoff": "2026-03-01", "contextWindow": 1000000, "maxOutput": 64000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.75, "outputPriceUSD": 3.75, "pricingNote": { "en": "Google lists the same introductory $0.75 / $3.75 pair as 3.8 Flash. Access requires Fairwind approval for trusted government, critical-infrastructure, and software-maintainer defenders. CyberGym and CWE-Bench figures on the launch post stay vendor-claimed.", "de": "Google listet dasselbe Einführungs-Paar $0,75 / $3,75 wie 3.8 Flash. Zugang erfordert Fairwind-Freigabe für vertrauenswürdige Regierungs-, Kritische-Infrastruktur- und Maintainer-Defender. CyberGym- und CWE-Bench-Zahlen im Launch-Post bleiben Vendor-Claims." }, "pricingChecked": "2026-09-05", "status": "solid", "sources": [ { "label": { "en": "Google Gemini 3.8 Flash Cyber announcement", "de": "Google-Gemini-3.8-Flash-Cyber-Ankündigung" }, "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/", "checkedAt": "2026-09-05" } ], "html": "https://verdictpal.com/models/gemini-3-8-flash-cyber" }, { "slug": "gemini-3-8-flash", "name": "Gemini 3.8 Flash", "provider": "Google", "family": "Gemini 3.8", "summary": { "en": "Google's 2 Sep 2026 Flash workhorse. Same intro API price as 3.7 Flash: $0.75 / $3.75 per 1M tokens through 31 Dec 2026. 1M context, 64K output.", "de": "Googles Flash-Arbeitspferd vom 2. Sep. 2026. Gleicher Intro-API-Preis wie 3.7 Flash: $0,75 / $3,75 pro 1M Tokens bis 31. Dez. 2026. 1M-Kontext, 64K-Output." }, "role": { "en": "Current Google Flash model for coding agents and high-throughput tool use. Prefer sourced AA ledger rows over launch DeepSWE / HLE-Verified / Vals Finance claims.", "de": "Aktuelles Google-Flash-Modell für Coding-Agents und Tool-Use mit hohem Durchsatz. Bevorzuge belegte AA-Ledger-Zeilen gegenüber Launch-Claims zu DeepSWE, HLE-Verified und Vals Finance." }, "releaseDate": "2026-09-02", "knowledgeCutoff": "2026-03-01", "contextWindow": 1000000, "maxOutput": 64000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.75, "outputPriceUSD": 3.75, "pricingNote": { "en": "Google introductory list through 31 Dec 2026: $0.75 input / $3.75 output per 1M tokens. From 1 Jan 2027 the stated list is $1.50 / $7.50. AA lists the same intro pair with a 90% cache-hit discount. Customizable effort levels. Restricted twin: Gemini 3.8 Flash Cyber (Fairwind).", "de": "Googles Einführungs-Liste bis 31. Dez. 2026: $0,75 Input / $3,75 Output pro 1M Tokens. Ab 1. Jan. 2027 gilt $1,50 / $7,50. AA führt dasselbe Intro-Paar mit 90 % Cache-Hit-Rabatt. Einstellbare Effort-Stufen. Eingeschränktes Zwilling: Gemini 3.8 Flash Cyber (Fairwind)." }, "pricingChecked": "2026-09-18", "status": "flagship", "sources": [ { "label": { "en": "Google Gemini 3.8 Flash announcement", "de": "Google-Gemini-3.8-Flash-Ankündigung" }, "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/", "checkedAt": "2026-09-18" }, { "label": { "en": "DeepMind Gemini 3.8 Flash model card", "de": "DeepMind-Modellkarte Gemini 3.8 Flash" }, "url": "https://deepmind.google/models/model-cards/gemini-3-8-flash/", "checkedAt": "2026-09-18" }, { "label": { "en": "Artificial Analysis Gemini 3.8 Flash", "de": "Artificial-Analysis-Gemini-3.8-Flash" }, "url": "https://artificialanalysis.ai/models/gemini-3-8-flash", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/gemini-3-8-flash" }, { "slug": "gemini-3-8-live-extended-thinking", "name": "Gemini 3.8 Live Extended Thinking", "provider": "Google", "family": "Gemini 3.8", "summary": { "en": "Google's 15 Sep 2026 high-reasoning Live API twin. Same 128K / 64K window and paid audio list as 3.8 Live ($3 / $12 per 1M tokens). Configurable thinking_level low/medium/high; async tools only. Knowledge cutoff January 2025.", "de": "Googles High-Reasoning-Live-API-Zwilling vom 15. Sep. 2026. Dasselbe 128K-/64K-Fenster und dieselbe bezahlte Audio-Liste wie 3.8 Live ($3 / $12 pro 1M Tokens). Konfigurierbares thinking_level low/medium/high; nur asynchrone Tools. Knowledge-Cutoff Januar 2025." }, "role": { "en": "Live voice agent for multi-step background reasoning and non-blocking tool calls while the spoken turn stays in motion.", "de": "Live-Voice-Agent für mehrstufiges Hintergrund-Reasoning und nicht-blockierende Tool-Calls, während der gesprochene Turn weiterläuft." }, "releaseDate": "2026-09-15", "knowledgeCutoff": "2025-01-01", "contextWindow": 131072, "maxOutput": 65536, "modalities": [ "text", "image", "audio", "video", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 12, "pricingNote": { "en": "Same Gemini API paid Live matrix as 3.8 Live (18 Sep 2026 docs): audio $3.00 / $12.00 per 1M tokens or $0.005 / $0.018 per minute; text $0.75 / $4.50; image/video input $1.00 per 1M. thinking_level supports low, medium, high — MINIMAL is not supported. Function calls must be NON_BLOCKING. Model ID gemini-3.8-live-extended-thinking.", "de": "Dieselbe Gemini-API Paid-Live-Matrix wie 3.8 Live (Docs 18. Sep. 2026): Audio $3,00 / $12,00 pro 1M Tokens oder $0,005 / $0,018 pro Minute; Text $0,75 / $4,50; Bild/Video-Input $1,00 pro 1M. thinking_level unterstützt low, medium, high — MINIMAL nicht. Function Calls müssen NON_BLOCKING sein. Modell-ID gemini-3.8-live-extended-thinking." }, "pricingChecked": "2026-09-18", "status": "solid", "sources": [ { "label": { "en": "Google Gemini 3.8 Live announcement", "de": "Google-Gemini-3.8-Live-Ankündigung" }, "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-live-gemini-3-8-live-extended-thinking/", "checkedAt": "2026-09-18" }, { "label": { "en": "Gemini API Extended Thinking model page", "de": "Gemini-API-Extended-Thinking-Modellseite" }, "url": "https://ai.google.dev/gemini-api/docs/models/gemini-3.8-live-extended-thinking", "checkedAt": "2026-09-18" }, { "label": { "en": "Gemini API pricing (Live)", "de": "Gemini-API-Preise (Live)" }, "url": "https://ai.google.dev/gemini-api/docs/pricing", "checkedAt": "2026-09-18" }, { "label": { "en": "DeepMind Gemini 3.8 Audio model card", "de": "DeepMind-Modellkarte Gemini 3.8 Audio" }, "url": "https://deepmind.google/models/model-cards/gemini-3-8-audio/", "checkedAt": "2026-09-18" }, { "label": { "en": "Developer Live API audio post", "de": "Developer-Post zur Live-API-Audio" }, "url": "https://blog.google/innovation-and-ai/technology/developers-tools/build-real-time-voice-applications-gemini-audio/", "checkedAt": "2026-09-18" } ], "html": "https://verdictpal.com/models/gemini-3-8-live-extended-thinking" }, { "slug": "gemini-3-8-live", "name": "Gemini 3.8 Live", "provider": "Google", "family": "Gemini 3.8", "summary": { "en": "Google's 15 Sep 2026 default Live API model for low-latency speech-to-speech. 128K context, 64K output. Paid audio list $3.00 / $12.00 per 1M tokens ($0.005 / $0.018 per minute). Text input/output is $0.75 / $4.50. Knowledge cutoff January 2025.", "de": "Googles Default-Live-API-Modell vom 15. Sep. 2026 für Speech-to-Speech mit niedriger Latenz. 128K-Kontext, 64K-Output. Bezahlte Audio-Liste $3,00 / $12,00 pro 1M Tokens ($0,005 / $0,018 pro Minute). Text-Input/Output $0,75 / $4,50. Knowledge-Cutoff Januar 2025." }, "role": { "en": "Current Google voice-agent workhorse for real-time dialogue, visual grounding, and async tool calls without configurable thinking_level.", "de": "Aktuelles Google-Voice-Agent-Arbeitspferd für Echtzeit-Dialog, visuelle Grounding und asynchrone Tool-Calls ohne konfigurierbares thinking_level." }, "releaseDate": "2026-09-15", "knowledgeCutoff": "2025-01-01", "contextWindow": 131072, "maxOutput": 65536, "modalities": [ "text", "image", "audio", "video", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 12, "pricingNote": { "en": "Gemini API paid Live rates (18 Sep 2026 docs): audio $3.00 input / $12.00 output per 1M tokens, or $0.005 / $0.018 per minute. Text $0.75 / $4.50. Image/video input $1.00 per 1M or $0.002/min. Caching not supported. Model ID gemini-3.8-live. Replaces gemini-3.1-flash-live-preview.", "de": "Gemini-API Paid-Live-Tarife (Docs 18. Sep. 2026): Audio $3,00 Input / $12,00 Output pro 1M Tokens, oder $0,005 / $0,018 pro Minute. Text $0,75 / $4,50. Bild/Video-Input $1,00 pro 1M oder $0,002/min. Caching nicht unterstützt. Modell-ID gemini-3.8-live. Ersetzt gemini-3.1-flash-live-preview." }, "pricingChecked": "2026-09-18", "status": "solid", "sources": [ { "label": { "en": "Google Gemini 3.8 Live announcement", "de": "Google-Gemini-3.8-Live-Ankündigung" }, "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-8-live-gemini-3-8-live-extended-thinking/", "checkedAt": "2026-09-18" }, { "label": { "en": "Gemini API model page", "de": "Gemini-API-Modellseite" }, "url": "https://ai.google.dev/gemini-api/docs/models/gemini-3.8-live", "checkedAt": "2026-09-18" }, { "label": { "en": "Gemini API pricing (Live)", "de": "Gemini-API-Preise (Live)" }, "url": "https://ai.google.dev/gemini-api/docs/pricing", "checkedAt": "2026-09-18" }, { "label": { "en": "DeepMind Gemini 3.8 Audio model card", "de": "DeepMind-Modellkarte Gemini 3.8 Audio" }, "url": "https://deepmind.google/models/model-cards/gemini-3-8-audio/", "checkedAt": "2026-09-18" } ], "html": "https://verdictpal.com/models/gemini-3-8-live" }, { "slug": "gemini-3-flash-reasoning", "name": "Gemini 3 Flash Preview (Reasoning)", "provider": "Google", "family": "Gemini 3", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.5/$3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.5/$3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-17", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.5, "outputPriceUSD": 3, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/gemini-3-flash-reasoning" }, { "slug": "gemini-3-flash", "name": "Gemini 3 Flash Preview", "provider": "Google", "family": "Gemini 3", "summary": { "en": "Google's December 2025 speed tier. 1M context, $0.50/$3 per 1M tokens. Pro-level reasoning at Flash pricing.", "de": "Googles Geschwindigkeits-Tier Dezember 2025. 1M-Kontext, $0,50/$3 pro 1M Tokens. Pro-Niveau-Reasoning zum Flash-Preis." }, "role": { "en": "Default Gemini route when 3.5 Flash is overkill and 3.1 Pro is too expensive. Strong on agentic coding at mid-tier cost.", "de": "Standard-Gemini-Route, wenn 3.5 Flash zu viel und 3.1 Pro zu teuer ist. Stark bei agentischem Coding zum Mittelklasse-Preis." }, "releaseDate": "2025-12-17", "knowledgeCutoff": "2025-01-31", "contextWindow": 1048576, "maxOutput": 65536, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.5, "outputPriceUSD": 3, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Gemini API pricing", "de": "Gemini-API-Preise" }, "url": "https://ai.google.dev/pricing", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-3-flash" }, { "slug": "gemini-3-pro-deep-think", "name": "Gemini 3 Pro Deep Think", "provider": "Google", "family": "Gemini 3", "summary": { "en": "Google's extended-reasoning Gemini 3 Pro variant. For hardest puzzle-style reasoning, not default chat.", "de": "Googles Extended-Reasoning-Variante von Gemini 3 Pro. Für härteste Puzzle-Reasoning-Aufgaben, nicht für Standard-Chat." }, "role": { "en": "Reasoning-heavy Gemini route when standard Pro isn't enough on ARC-style tasks.", "de": "Reasoning-lastige Gemini-Route, wenn Standard-Pro bei ARC-ähnlichen Aufgaben nicht reicht." }, "releaseDate": "2025-12-01", "knowledgeCutoff": "2025-01-31", "contextWindow": 1048576, "maxOutput": 65536, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Custom / preview pricing — check Google AI Studio for current rates.", "de": "Individuelle / Preview-Preise — aktuelle Tarife in Google AI Studio prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "ARC Prize leaderboard", "de": "ARC-Prize-Leaderboard" }, "url": "https://arcprize.org/leaderboard", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-3-pro-deep-think" }, { "slug": "gemini-3-pro-low", "name": "Gemini 3 Pro Preview (low)", "provider": "Google", "family": "Gemini 3", "summary": { "en": "Listed on Artificial Analysis with API pricing $2/$12 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $2/$12 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-11-18", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 2, "outputPriceUSD": 12, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/gemini-3-pro-low" }, { "slug": "gemini-3-pro", "name": "Gemini 3 Pro", "provider": "Google", "family": "Gemini 3", "summary": { "en": "Google's November 2025 Pro-tier Gemini 3. LegalBench 87% on Vals. Superseded by 3.1 Pro for most frontier work but still on API.", "de": "Googles Pro-Tier Gemini 3 November 2025. LegalBench 87 % auf Vals. Für die meiste Frontier-Arbeit von 3.1 Pro abgelöst, aber weiterhin per API verfügbar." }, "role": { "en": "Prior-generation Pro Gemini for workflows already pinned to gemini-3-pro endpoints.", "de": "Vorgänger-Pro-Gemini für Workflows, die bereits an gemini-3-pro-Endpunkte gebunden sind." }, "releaseDate": "2025-11-18", "knowledgeCutoff": "2025-01-31", "contextWindow": 1048576, "maxOutput": 65536, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 12, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Gemini 3 docs", "de": "Gemini-3-Docs" }, "url": "https://ai.google.dev/gemini-api/docs/gemini-3", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemini-3-pro" }, { "slug": "gemma-4-12b-non-reasoning", "name": "Gemma 4 12B (Non-reasoning)", "provider": "Google", "family": "Gemma", "summary": { "en": "Google's compact Gemma 4 non-reasoning variant, evaluated by Artificial Analysis on 10 Jun. The tool is intentionally light until benchmark rows are imported into the local ledger.", "de": "Googles kompakte Gemma-4-Non-Reasoning-Variante, am 10. Juni von Artificial Analysis evaluiert. Das Tool bleibt bewusst leicht, bis Benchmark-Zeilen ins lokale Ledger importiert sind." }, "role": { "en": "Cheap local baseline for tagging, extraction, and latency-sensitive prompts where explicit reasoning traces are unnecessary.", "de": "Günstige lokale Baseline für Tagging, Extraktion und latenzsensitive Prompts, bei denen explizite Reasoning-Traces unnötig sind." }, "releaseDate": "2026-06-10", "knowledgeCutoff": "open", "contextWindow": 128000, "maxOutput": 8192, "modalities": [ "text", "code" ], "access": [ "open-weights", "on-device" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Open weights — no API list price; hosting cost depends on your GPU provider.", "de": "Open Weights — kein API-Listenpreis; Hosting-Kosten hängen vom GPU-Anbieter ab." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Gemma docs", "de": "Gemma-Docs" }, "url": "https://ai.google.dev/gemma", "checkedAt": "2026-06-14" }, { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/gemma-4-12b-non-reasoning" }, { "slug": "gemma-4-12b-reasoning", "name": "Gemma 4 12B (Reasoning)", "provider": "Google", "family": "Gemma", "summary": { "en": "Google's smaller Gemma 4 reasoning variant, evaluated by Artificial Analysis on 7 Jun. Treat it as a compact open-weights reasoning candidate until the local benchmark ledger has sourced rows.", "de": "Googles kleinere Gemma-4-Reasoning-Variante, am 7. Juni von Artificial Analysis evaluiert. Bis das lokale Benchmark-Ledger belegte Zeilen hat, als kompakten Open-Weights-Reasoning-Kandidaten behandeln." }, "role": { "en": "Small open-weights reasoning model for local experiments, classroom demos, and routing tests before paying for frontier APIs.", "de": "Kleines Open-Weights-Reasoning-Modell für lokale Experimente, Unterrichtsdemos und Routing-Tests vor teuren Frontier-APIs." }, "releaseDate": "2026-06-07", "knowledgeCutoff": "open", "contextWindow": 128000, "maxOutput": 8192, "modalities": [ "text", "code", "reasoning" ], "access": [ "open-weights", "on-device" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Open weights — no API list price; hosting cost depends on your GPU provider.", "de": "Open Weights — kein API-Listenpreis; Hosting-Kosten hängen vom GPU-Anbieter ab." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Gemma docs", "de": "Gemma-Docs" }, "url": "https://ai.google.dev/gemma", "checkedAt": "2026-06-14" }, { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/gemma-4-12b-reasoning" }, { "slug": "gemma-4-31b", "name": "Gemma 4 31B IT", "provider": "Google", "family": "Gemma", "summary": { "en": "Google's March 2026 open-weights Gemma 4 instruct model. SAGE 55% on Vals (#2). Small enough to self-host on a single GPU cluster.", "de": "Googles Open-Weights-Gemma-4-Instruct-Modell März 2026. SAGE 55 % auf Vals (#2). Klein genug für Self-Hosting auf einem GPU-Cluster." }, "role": { "en": "Self-hosted Google open weights for privacy-sensitive workflows that can't call cloud APIs.", "de": "Selbst gehostete Google-Open-Weights für datenschutzsensible Workflows ohne Cloud-APIs." }, "releaseDate": "2026-03-01", "knowledgeCutoff": "open", "contextWindow": 131072, "maxOutput": 8192, "modalities": [ "text", "code" ], "access": [ "open-weights", "on-device" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Open weights — no API list price; host yourself or use a cloud GPU provider.", "de": "Open Weights — kein API-Listenpreis; selbst hosten oder Cloud-GPU-Anbieter nutzen." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Gemma docs", "de": "Gemma-Docs" }, "url": "https://ai.google.dev/gemma", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gemma-4-31b" }, { "slug": "glm-4-7", "name": "GLM-4.7 (Reasoning)", "provider": "Zhipu", "family": "GLM-4.7 (Reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.6/$2.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.6/$2.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-22", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.6, "outputPriceUSD": 2.2, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/glm-4-7" }, { "slug": "glm-5-1", "name": "GLM-5.1", "provider": "Zhipu", "family": "GLM", "summary": { "en": "Z.ai's April 2026 flagship for long-horizon agentic engineering. Official docs list text input/output, 200K context, 128K maximum output, function calling, structured output, context caching, MCP, and open weights under the MIT license.", "de": "Z.ais April-2026-Flaggschiff für Long-Horizon-Agentic-Engineering. Offizielle Docs nennen Text-Input/-Output, 200K Kontext, 128K Max-Output, Function Calling, Structured Output, Context Caching, MCP und Open Weights unter MIT-Lizenz." }, "role": { "en": "Long-horizon coding and autonomous agent model for Claude Code/OpenClaw-style workflows when you want open weights from the GLM line.", "de": "Long-Horizon-Coding- und autonome-Agenten-Modell für Claude-Code-/OpenClaw-artige Workflows, wenn Open Weights aus der GLM-Linie gefragt sind." }, "releaseDate": "2026-04-07", "knowledgeCutoff": "open", "contextWindow": 200000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Available through api.z.ai, BigModel.cn, Z.ai Coding Plan, Hugging Face, and ModelScope. Coding Plan usage consumes quota at model-dependent multipliers; no stable USD API list price captured in the local ledger yet.", "de": "Verfügbar über api.z.ai, BigModel.cn, Z.ai Coding Plan, Hugging Face und ModelScope. Coding-Plan-Nutzung verbraucht Quote mit modellabhängigen Multiplikatoren; noch kein stabiler USD-API-Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "solid", "sources": [ { "label": { "en": "GLM-5.1 docs", "de": "GLM-5.1-Dokumentation" }, "url": "https://docs.z.ai/guides/llm/glm-5.1", "checkedAt": "2026-06-14" }, { "label": { "en": "GLM-5.1 announcement", "de": "GLM-5.1-Ankündigung" }, "url": "https://z.ai/blog/glm-5.1", "checkedAt": "2026-06-14" }, { "label": { "en": "GLM-5.1 on Hugging Face", "de": "GLM-5.1 auf Hugging Face" }, "url": "https://huggingface.co/zai-org/GLM-5.1", "checkedAt": "2026-06-14" }, { "label": { "en": "Artificial Analysis comparison", "de": "Artificial-Analysis-Vergleich" }, "url": "https://artificialanalysis.ai/models/comparisons/kimi-k2-6-vs-glm-5-1-non-reasoning", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/glm-5-1" }, { "slug": "glm-5-2-non-reasoning", "name": "GLM-5.2 (Non-reasoning)", "provider": "Zhipu", "family": "GLM-5.2 (Non-reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.351/$4.25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.351/$4.25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-06-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 1.351, "outputPriceUSD": 4.25, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/glm-5-2-non-reasoning" }, { "slug": "glm-5-2", "name": "GLM-5.2", "provider": "Zhipu", "family": "GLM", "summary": { "en": "Highest-ranked open-source model on FrontierSWE, PostTrainBench, and SWE-Marathon. $1.40 / $4.40 per 1M tokens, MIT weights pending.", "de": "Höchstplatziertes Open-Source-Modell auf FrontierSWE, PostTrainBench und SWE-Marathon. $1,40 / $4,40 pro 1M Tokens, MIT-Weights ausstehend." }, "role": { "en": "Long-horizon coding and autonomous agent model for project-scale engineering. The GLM pick when you need 1M context that stays usable across multi-day refactors, not just a large window.", "de": "Long-Horizon-Coding- und autonomes-Agenten-Modell für projektgrosse Engineering-Aufgaben. Die GLM-Wahl, wenn 1M-Kontext gebraucht wird, der über mehrtägige Refactors nutzbar bleibt, nicht nur ein grosses Fenster." }, "releaseDate": "2026-06-13", "knowledgeCutoff": "open", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 1.4, "outputPriceUSD": 4.4, "pricingNote": { "en": "API at $1.40 / $4.40 per 1M tokens. MIT open weights announced for the week of June 16. GLM Coding Plan from $18/mo includes quota across GLM models.", "de": "API bei $1,40 / $4,40 pro 1M Tokens. MIT-Open-Weights für die Woche ab 16. Juni angekündigt. GLM Coding Plan ab $18/Monat inklusive Quote über GLM-Modelle." }, "pricingChecked": "2026-06-19", "status": "solid", "sources": [ { "label": { "en": "GLM-5.2 docs", "de": "GLM-5.2-Dokumentation" }, "url": "https://docs.z.ai/guides/llm/glm-5.2", "checkedAt": "2026-06-19" }, { "label": { "en": "Z.ai pricing", "de": "Z.ai-Preise" }, "url": "https://docs.z.ai/guides/overview/pricing", "checkedAt": "2026-06-19" }, { "label": { "en": "Artificial Analysis — GLM-5.2 leading open weights", "de": "Artificial Analysis — GLM-5.2 führende Open Weights" }, "url": "https://artificialanalysis.ai/articles/glm-5-2-is-the-new-leading-open-weights-model", "checkedAt": "2026-06-19" }, { "label": { "en": "Requesty API pricing", "de": "Requesty-API-Preise" }, "url": "https://www.requesty.ai/models/zai/glm-5.2", "checkedAt": "2026-06-19" } ], "html": "https://verdictpal.com/models/glm-5-2" }, { "slug": "glm-5-3-flash", "name": "GLM-5.3-Flash", "provider": "Zhipu", "family": "GLM", "summary": { "en": "Z.ai's 26 Aug 2026 Flash sibling of GLM-5.3.", "de": "Z.ais Flash-Geschwister von GLM-5.3 vom 26. Aug. 2026." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-26", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights" ], "inputPriceUSD": 0.15, "outputPriceUSD": 0.5, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Z.ai GLM-5.3-Flash announcement", "de": "Z.ai-GLM-5.3-Flash-Ankündigung" }, "url": "https://z.ai/blog/glm-5.3-flash", "checkedAt": "2026-09-01" }, { "label": { "en": "Hugging Face GLM-5.3-Flash", "de": "Hugging-Face-GLM-5.3-Flash" }, "url": "https://huggingface.co/zai-org/GLM-5.3-Flash", "checkedAt": "2026-09-01" }, { "label": { "en": "Artificial Analysis GLM-5.3-Flash", "de": "Artificial-Analysis-GLM-5.3-Flash" }, "url": "https://artificialanalysis.ai/models/glm-5-3-flash", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/glm-5-3-flash" }, { "slug": "glm-5-3", "name": "GLM-5.3", "provider": "Zhipu", "family": "GLM", "summary": { "en": "Z.ai's August 2026 coding refresh of the GLM-5.2 base. API list price $1.40 / $4.40 per 1M tokens — same slot as GLM-5.2.", "de": "Z.ais Coding-Refresh August 2026 auf der GLM-5.2-Basis. API-Listenpreis $1,40 / $4,40 pro 1M Tokens — derselbe Slot wie GLM-5.2." }, "role": { "en": "Current Z.ai coding-plan and API model for long-horizon agent work. Prefer sourced ledger rows over vendor coding/cyber tables; AA scores sync from the public snapshot. Open weights were promised after a safety delay, not at launch.", "de": "Aktuelles Z.ai-Coding-Plan- und API-Modell für Long-Horizon-Agentenarbeit. Bevorzuge belegte Ledger-Zeilen gegenüber Vendor-Coding-/Cyber-Tabellen; AA-Scores kommen aus dem öffentlichen Snapshot. Open Weights waren nach einem Safety-Delay angekündigt, nicht zum Launch." }, "releaseDate": "2026-08-14", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 1.4, "outputPriceUSD": 4.4, "pricingNote": { "en": "Z.ai API list as of 2026-08-24, matching AA: $1.40 input / $4.40 output per 1M tokens; cached input listed at $0.26. GLM Coding Plan is a separate points/quota subscription, not this token pair. Weights were not public on this check.", "de": "Z.ai-API-Liste Stand 2026-08-24, gleich AA: $1,40 Input / $4,40 Output pro 1M Tokens; Cache-Input mit $0,26. Der GLM Coding Plan ist ein separates Punkte-/Quota-Abo, nicht dieses Token-Paar. Weights waren bei dieser Prüfung nicht öffentlich." }, "pricingChecked": "2026-08-24", "status": "solid", "sources": [ { "label": { "en": "Z.ai GLM-5.3 announcement", "de": "Z.ai-GLM-5.3-Ankündigung" }, "url": "https://z.ai/blog/glm-5.3", "checkedAt": "2026-08-24" }, { "label": { "en": "Z.ai pricing", "de": "Z.ai-Preise" }, "url": "https://docs.z.ai/guides/overview/pricing", "checkedAt": "2026-08-24" }, { "label": { "en": "Artificial Analysis GLM-5.3", "de": "Artificial-Analysis-GLM-5.3" }, "url": "https://artificialanalysis.ai/models/glm-5-3", "checkedAt": "2026-08-24" } ], "html": "https://verdictpal.com/models/glm-5-3" }, { "slug": "glm-5-non-reasoning", "name": "GLM-5 (Non-reasoning)", "provider": "Zhipu", "family": "GLM-5 (Non-reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1/$3.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1/$3.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-11", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 1, "outputPriceUSD": 3.2, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/glm-5-non-reasoning" }, { "slug": "glm-5-turbo", "name": "GLM-5-Turbo", "provider": "Zhipu", "family": "Zhipu", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-15", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/glm-5-turbo" }, { "slug": "glm-5", "name": "GLM-5 (Reasoning)", "provider": "Zhipu", "family": "GLM-5 (Reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1/$3.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1/$3.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-11", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1, "outputPriceUSD": 3.2, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/glm-5" }, { "slug": "glm-5v-turbo", "name": "GLM 5V Turbo (Reasoning)", "provider": "Zhipu", "family": "GLM 5V", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-01", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/glm-5v-turbo" }, { "slug": "gpt-4-2023", "name": "GPT-4 (2023)", "provider": "OpenAI", "family": "GPT-4", "summary": { "en": "Original GPT-4 technical report baseline. Anchor for pre-2024 frontier comparisons.", "de": "Original-GPT-4-Technical-Report-Baseline. Anker für Frontier-Vergleiche vor 2024." }, "role": { "en": "Historical anchor only — not deployable on current OpenAI routes.", "de": "Nur historischer Anker — nicht auf aktuellen OpenAI-Routen deploybar." }, "releaseDate": "2023-03-14", "knowledgeCutoff": "2021-09-01", "contextWindow": 8192, "maxOutput": 4096, "modalities": [ "text", "code" ], "access": [ "api" ], "inputPriceUSD": 30, "outputPriceUSD": 60, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "GPT-4 technical report", "de": "GPT-4-Technical-Report" }, "url": "https://arxiv.org/abs/2303.08774", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gpt-4-2023" }, { "slug": "gpt-4o-2024-05", "name": "GPT-4o (2024-05)", "provider": "OpenAI", "family": "GPT-4o", "summary": { "en": "OpenAI's May 2024 GPT-4o flagship. The reference GPT-4o checkpoint before GPT-5.", "de": "OpenAIs GPT-4o-Flaggschiff Mai 2024. Referenz-GPT-4o-Checkpoint vor GPT-5." }, "role": { "en": "Legacy OpenAI multimodal baseline — use GPT-5.4/5.5 for new work.", "de": "Legacy-OpenAI-Multimodal-Baseline — für neue Arbeit GPT-5.4/5.5 nutzen." }, "releaseDate": "2024-05-13", "knowledgeCutoff": "2023-10-01", "contextWindow": 128000, "maxOutput": 16384, "modalities": [ "text", "image", "audio", "code", "tool-use" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2.5, "outputPriceUSD": 10, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "GPT-4o announcement", "de": "GPT-4o-Ankündigung" }, "url": "https://openai.com/index/hello-gpt-4o/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gpt-4o-2024-05" }, { "slug": "gpt-4o-2024", "name": "GPT-4o (2024)", "provider": "OpenAI", "family": "GPT-4o", "summary": { "en": "Early GPT-4o API snapshot. Useful only for comparing against May 2024+ checkpoints.", "de": "Früher GPT-4o-API-Snapshot. Nur zum Vergleich mit Checkpoints ab Mai 2024 nützlich." }, "role": { "en": "Earliest GPT-4o ledger row — prefer gpt-4o-2024-05 for baseline comparisons.", "de": "Früheste GPT-4o-Ledger-Zeile — für Baseline-Vergleiche gpt-4o-2024-05 bevorzugen." }, "releaseDate": "2024-05-01", "knowledgeCutoff": "2023-10-01", "contextWindow": 128000, "maxOutput": 16384, "modalities": [ "text", "image", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 5, "outputPriceUSD": 15, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "GPT-4o announcement", "de": "GPT-4o-Ankündigung" }, "url": "https://openai.com/index/hello-gpt-4o/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gpt-4o-2024" }, { "slug": "gpt-4v-2023", "name": "GPT-4V (2023)", "provider": "OpenAI", "family": "GPT-4", "summary": { "en": "OpenAI's September 2023 vision-capable GPT-4. Early multimodal baseline on public leaderboards before GPT-4o.", "de": "OpenAIs vision-fähiges GPT-4 September 2023. Frühe Multimodal-Baseline auf öffentlichen Leaderboards vor GPT-4o." }, "role": { "en": "Historical vision baseline only — use GPT-4o or GPT-5 for current multimodal work.", "de": "Nur historische Vision-Baseline — für aktuelle Multimodal-Arbeit GPT-4o oder GPT-5 nutzen." }, "releaseDate": "2023-09-25", "knowledgeCutoff": "2021-09-01", "contextWindow": 128000, "maxOutput": 4096, "modalities": [ "text", "image", "code" ], "access": [ "api" ], "inputPriceUSD": 10, "outputPriceUSD": 30, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "GPT-4V system card", "de": "GPT-4V-Systemkarte" }, "url": "https://openai.com/research/gpt-4v-system-card", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gpt-4v-2023" }, { "slug": "gpt-5-1-codex", "name": "GPT-5.1 Codex (high)", "provider": "OpenAI", "family": "GPT-5.1 Codex", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$10 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$10 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-11-13", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 10, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/gpt-5-1-codex" }, { "slug": "gpt-5-1", "name": "GPT-5.1", "provider": "OpenAI", "family": "GPT-5", "summary": { "en": "OpenAI's November 2025 GPT-5.1 family. $1.25/$10 per 1M tokens, 400K context. Retired from ChatGPT March 2026; API snapshots may still exist. MedScribe leader on Vals (88.1%).", "de": "OpenAIs GPT-5.1-Familie November 2025. $1,25/$10 pro 1M Tokens, 400K-Kontext. Aus ChatGPT ausgemustert März 2026; API-Snapshots können noch existieren. MedScribe-Spitze auf Vals (88,1 %)." }, "role": { "en": "Legacy GPT-5.1 endpoints only — migrate to GPT-5.4/5.5 for new work.", "de": "Nur Legacy-GPT-5.1-Endpunkte — für neue Arbeit auf GPT-5.4/5.5 migrieren." }, "releaseDate": "2025-11-13", "knowledgeCutoff": "2024-09-30", "contextWindow": 400000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 1.25, "outputPriceUSD": 10, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "OpenAI GPT-5.1 docs", "de": "OpenAI-GPT-5.1-Docs" }, "url": "https://developers.openai.com/api/docs/models/gpt-5.1", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gpt-5-1" }, { "slug": "gpt-5-2-codex", "name": "GPT-5.2 Codex (xhigh)", "provider": "OpenAI", "family": "GPT-5.2 Codex", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.75/$14 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.75/$14 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-11", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.75, "outputPriceUSD": 14, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/gpt-5-2-codex" }, { "slug": "gpt-5-2-medium", "name": "GPT-5.2 (medium)", "provider": "OpenAI", "family": "GPT-5.2 (medium)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.75/$14 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.75/$14 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-11", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.75, "outputPriceUSD": 14, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/gpt-5-2-medium" }, { "slug": "gpt-5-2-pro", "name": "GPT-5.2 Pro", "provider": "OpenAI", "family": "GPT-5", "summary": { "en": "OpenAI's premium GPT-5.2 tier. Higher reasoning ceiling than GPT-5.2 standard.", "de": "OpenAIs Premium-GPT-5.2-Tier. Höhere Reasoning-Decke als Standard-GPT-5.2." }, "role": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger." }, "releaseDate": "2026-05-12", "knowledgeCutoff": "2025-08-01", "contextWindow": 410000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Pro-tier pricing — check OpenAI API pricing page for current rates.", "de": "Pro-Tier-Preise — aktuelle Tarife auf der OpenAI-API-Preisseite prüfen." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "OpenAI GDPval (arXiv)", "de": "OpenAI-GDPval (arXiv)" }, "url": "https://arxiv.org/abs/2510.04374", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/gpt-5-2-pro" }, { "slug": "gpt-5-2", "name": "GPT-5.2", "provider": "OpenAI", "family": "GPT-5", "summary": { "en": "OpenAI's December 2025 frontier. 410K context, 128K output, configurable reasoning effort. $1.75/$14 per 1M tokens.", "de": "Openais Frontier-Modell Dezember 2025. 410K-Kontext, 128K-Output, konfigurierbarer Reasoning-Aufwand. $1,75/$14 pro 1M Tokens." }, "role": { "en": "Previous-gen GPT-5 for complex professional work. Superseded by GPT-5.4/5.5 but still available at lower cost.", "de": "Vorherige GPT-5-Generation für komplexe professionelle Arbeit. Durch GPT-5.4/5.5 abgelöst, aber noch verfügbar zum niedrigeren Preis." }, "releaseDate": "2025-12-11", "knowledgeCutoff": "2025-08-01", "contextWindow": 410000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 1.75, "outputPriceUSD": 14, "pricingNote": { "en": "Cached input at 10% of normal price. Batch API cuts costs 50%.", "de": "Gecachter Input zum 10 % des Normalpreises. Batch-API reduziert Kosten um 50 %." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "OpenAI announcement", "de": "OpenAI-Ankündigung" }, "url": "https://openai.com/index/introducing-gpt-5-2/", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/gpt-5-2" }, { "slug": "gpt-5-3-codex", "name": "GPT-5.3 Codex (xhigh)", "provider": "OpenAI", "family": "GPT-5.3 Codex", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.75/$14 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.75/$14 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-05", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.75, "outputPriceUSD": 14, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/gpt-5-3-codex" }, { "slug": "gpt-5-4-low", "name": "GPT-5.4 (low)", "provider": "OpenAI", "family": "GPT-5.4 (low)", "summary": { "en": "Listed on Artificial Analysis with API pricing $2.5/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $2.5/$15 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-05", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 2.5, "outputPriceUSD": 15, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/gpt-5-4-low" }, { "slug": "gpt-5-4-mini", "name": "GPT-5.4 Mini", "provider": "OpenAI", "family": "GPT-5.4", "summary": { "en": "OpenAI's March 2026 mid-tier. Best price/performance pick for production apps. $0.75/$4.50 per 1M tokens.", "de": "Openais Mittelklasse März 2026. Bestes Preis-/Leistungs-Verhältnis für Produktions-Apps. $0,75/$4,50 pro 1M Tokens." }, "role": { "en": "Cost-effective GPT-5.4 for production chatbots, code completions, and real-time applications.", "de": "Kostengünstiges GPT-5.4 für Produktions-Chatbots, Code-Ergänzungen und Echtzeit-Anwendungen." }, "releaseDate": "2026-03-17", "knowledgeCutoff": "2026-01-01", "contextWindow": 128000, "maxOutput": 16000, "modalities": [ "text", "image", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.75, "outputPriceUSD": 4.5, "pricingNote": { "en": "Available to free-tier users. Cached input at 10% of normal price.", "de": "Für Free-Tier-Nutzer verfügbar. Gecachter Input zum 10 % des Normalpreises." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "OpenAI announcement", "de": "OpenAI-Ankündigung" }, "url": "https://openai.com/index/introducing-gpt-5-4/", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/gpt-5-4-mini" }, { "slug": "gpt-5-4-nano", "name": "GPT-5.4 Nano", "provider": "OpenAI", "family": "GPT-5.4", "summary": { "en": "OpenAI's March 2026 budget tier. $0.20/$1.25 per 1M tokens. Optimized for speed-critical, high-volume tasks.", "de": "Openais Budget-Stufe März 2026. $0,20/$1,25 pro 1M Tokens. Optimiert für geschwindigkeitskritische, hochvolumige Aufgaben." }, "role": { "en": "Cheapest GPT-5.4 for routing, extraction, classification, and high-volume backend tasks.", "de": "Günstigstes GPT-5.4 für Routing, Extraktion, Klassifikation und hochvolumige Backend-Aufgaben." }, "releaseDate": "2026-03-17", "knowledgeCutoff": "2026-01-01", "contextWindow": 128000, "maxOutput": 16000, "modalities": [ "text", "image", "code", "tool-use" ], "access": [ "api" ], "inputPriceUSD": 0.2, "outputPriceUSD": 1.25, "pricingNote": { "en": "API-only. Cached input at 10% of normal price. 4x more expensive than GPT-5 nano equivalent.", "de": "Nur API. Gecachter Input zum 10 % des Normalpreises. 4× teurer als GPT-5-Nano-Äquivalent." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "OpenAI announcement", "de": "OpenAI-Ankündigung" }, "url": "https://openai.com/index/introducing-gpt-5-4/", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/gpt-5-4-nano" }, { "slug": "gpt-5-4-pro", "name": "GPT-5.4 Pro", "provider": "OpenAI", "family": "GPT-5.4", "summary": { "en": "OpenAI's March 2026 premium tier. $30/$180 per 1M tokens. Highest-precision variant for expensive-but-important workloads.", "de": "Openais Premium-Stufe März 2026. $30/$180 pro 1M Tokens. Höchstpräzise Variante für teure, aber wichtige Workloads." }, "role": { "en": "Highest-precision GPT-5.4 for financial analysis, scientific research, and high-stakes decision support.", "de": "Höchstpräzises GPT-5.4 für Finanzanalyse, wissenschaftliche Forschung und Entscheidungsunterstützung mit hohem Einsatz." }, "releaseDate": "2026-03-05", "knowledgeCutoff": "2026-01-01", "contextWindow": 128000, "maxOutput": 32000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 30, "outputPriceUSD": 180, "pricingNote": { "en": "Premium pricing. No cached input discount. 12x more expensive than base GPT-5.4.", "de": "Premium-Preise. Kein Rabatt für gecachten Input. 12× teurer als GPT-5.4-Basis." }, "pricingChecked": "2026-06-02", "status": "solid", "sources": [ { "label": { "en": "OpenAI announcement", "de": "OpenAI-Ankündigung" }, "url": "https://openai.com/index/introducing-gpt-5-4/", "checkedAt": "2026-06-02" } ], "html": "https://verdictpal.com/models/gpt-5-4-pro" }, { "slug": "gpt-5-4", "name": "GPT 5.4", "provider": "OpenAI", "family": "GPT 5", "summary": { "en": "OpenAI's March 2026 model.", "de": "OpenAIs Modell März 2026." }, "role": { "en": "Default for production OpenAI workloads where GPT 5.5 is overkill but GPT 5.2 is too old.", "de": "Standard für OpenAI-Produktions-Workloads, bei denen GPT 5.5 zu viel, GPT 5.2 aber zu alt ist." }, "releaseDate": "2026-03-05", "knowledgeCutoff": "2026-01-31", "contextWindow": 256000, "maxOutput": 65536, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 7, "outputPriceUSD": 28, "pricingNote": { "en": "44% cheaper than GPT 5.5 on input. Batch API and prompt caching cut effective list price further.", "de": "44 % günstiger als GPT 5.5 beim Input. Batch-API und Prompt-Caching senken den effektiven Listenpreis weiter." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "OpenAI pricing", "de": "OpenAI-Preise" }, "url": "https://openai.com/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "GPT 5.4 release", "de": "GPT-5.4-Release" }, "url": "https://openai.com/gpt-5-4", "checkedAt": "2026-06-01" }, { "label": { "en": "Artificial Analysis", "de": "Artificial Analysis" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/gpt-5-4" }, { "slug": "gpt-5-5-high", "name": "GPT-5.5 (high)", "provider": "OpenAI", "family": "GPT-5.5 (high)", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$30 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$30 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 30, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/gpt-5-5-high" }, { "slug": "gpt-5-5-instant-05-26", "name": "GPT-5.5 Instant (May 2026)", "provider": "OpenAI", "family": "GPT-5.5 Instant", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$30 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$30 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-05", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 30, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/gpt-5-5-instant-05-26" }, { "slug": "gpt-5-5-instant-06-26", "name": "GPT-5.5 Instant (June 2026)", "provider": "OpenAI", "family": "GPT-5.5 Instant", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$30 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$30 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-06-25", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 30, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/gpt-5-5-instant-06-26" }, { "slug": "gpt-5-5-low", "name": "GPT-5.5 (low)", "provider": "OpenAI", "family": "GPT-5.5 (low)", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$30 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$30 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 30, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/gpt-5-5-low" }, { "slug": "gpt-5-5-medium", "name": "GPT-5.5 (medium)", "provider": "OpenAI", "family": "GPT-5.5 (medium)", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$30 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$30 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 30, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/gpt-5-5-medium" }, { "slug": "gpt-5-5-non-reasoning", "name": "GPT-5.5 (Non-reasoning)", "provider": "OpenAI", "family": "GPT-5.5 (Non-reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $5/$30 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $5/$30 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 5, "outputPriceUSD": 30, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/gpt-5-5-non-reasoning" }, { "slug": "gpt-5-5-pro", "name": "GPT-5.5 Pro", "provider": "OpenAI", "family": "GPT 5", "summary": { "en": "OpenAI's premium GPT-5.5 tier. Responses API only — higher reasoning ceiling than standard GPT-5.5.", "de": "OpenAIs Premium-GPT-5.5-Tier. Nur Responses API — höhere Reasoning-Decke als Standard-GPT-5.5." }, "role": { "en": "OpenAI Pro-tier pick when accuracy matters more than latency — legal, finance, and multi-pass research workflows.", "de": "OpenAI-Pro-Tier-Wahl, wenn Genauigkeit wichtiger ist als Latenz — Recht, Finanzen und mehrstufige Research-Workflows." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "2025-12-01", "contextWindow": 1050000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 30, "outputPriceUSD": 180, "pricingNote": { "en": "No cached-input discount. Long-context tier ($60 / $270 per 1M) above 272K input tokens. Batch API at 50% off.", "de": "Kein Rabatt für gecachten Input. Long-Context-Tier ($60 / $270 pro 1M) über 272K Input-Tokens. Batch-API mit 50 % Rabatt." }, "pricingChecked": "2026-06-10", "status": "solid", "sources": [ { "label": { "en": "OpenAI announcement", "de": "OpenAI-Ankündigung" }, "url": "https://openai.com/index/introducing-gpt-5-5/", "checkedAt": "2026-06-10" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/models/gpt-5.5-pro", "checkedAt": "2026-06-10" } ], "html": "https://verdictpal.com/models/gpt-5-5-pro" }, { "slug": "gpt-5-5", "name": "GPT 5.5", "provider": "OpenAI", "family": "GPT 5", "summary": { "en": "OpenAI's April 2026 flagship.", "de": "OpenAIs Flaggschiff April 2026." }, "role": { "en": "Best generalist for coding agents, research workflows, and tool-heavy automation. The model the desk reaches for when we need coding agents that close the loop on a single issue.", "de": "Bester Generalist für Coding-Agents, Research-Workflows und tool-lastige Automatisierung. Das Modell, zu dem der Desk greift, wenn Coding-Agents ein Issue in einem Durchlauf schliessen sollen." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "2026-03-15", "contextWindow": 512000, "maxOutput": 128000, "modalities": [ "text", "image", "audio", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 12.5, "outputPriceUSD": 50, "pricingNote": { "en": "Batch API and prompt caching cut effective list price up to 75%.", "de": "Batch-API und Prompt-Caching senken den Listenpreis effektiv um bis zu 75 %." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "OpenAI pricing", "de": "OpenAI-Preise" }, "url": "https://openai.com/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "GPT 5 family announcement", "de": "GPT-5-Family-Ankündigung" }, "url": "https://openai.com/gpt-5", "checkedAt": "2026-06-01" }, { "label": { "en": "Artificial Analysis", "de": "Artificial Analysis" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/gpt-5-5" }, { "slug": "gpt-5-6-luna", "name": "GPT-5.6 Luna", "provider": "OpenAI", "family": "GPT-5.6", "summary": { "en": "OpenAI's generally available GPT-5.6 Luna line for high-volume work. API list for short context, checked 23 Sep 2026, is $0.20 / $1.20 per 1M tokens. GPT-6 Luna is the current Luna tier.", "de": "OpenAIs allgemein verfügbare GPT-5.6-Luna-Linie für High-Volume-Arbeit. Die API-Liste für kurzen Kontext, geprüft am 23. Sep. 2026, ist $0,20 / $1,20 pro 1M Tokens. GPT-6 Luna ist die aktuelle Luna-Stufe." }, "role": { "en": "GPT-5.6 tier for high-volume sub-agent, extraction, routing, and everyday coding workloads.", "de": "GPT-5.6-Stufe für High-Volume-Sub-Agenten, Extraktion, Routing und alltägliche Coding-Workloads." }, "releaseDate": "2026-07-09", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.2, "outputPriceUSD": 1.2, "pricingNote": { "en": "OpenAI API standard list checked 23 Sep 2026: $0.20 input / $1.20 output per 1M tokens on short context. Cached input $0.02; cache writes $0.25. Context limit is still unpublished. GPT-6 Luna is the current Luna tier.", "de": "OpenAI-API-Standardliste geprüft am 23. Sep. 2026: $0,20 Input / $1,20 Output pro 1M Tokens bei kurzem Kontext. Cached Input $0,02; Cache-Writes $0,25. Das Kontextlimit ist weiter unveröffentlicht. GPT-6 Luna ist die aktuelle Luna-Stufe." }, "pricingChecked": "2026-09-23", "status": "solid", "sources": [ { "label": { "en": "OpenAI GPT-5.6 launch", "de": "OpenAI-GPT-5.6-Launch" }, "url": "https://openai.com/index/gpt-5-6/", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI GPT-5.6 preview availability and pricing", "de": "OpenAI GPT-5.6-Preview-Verfügbarkeit und Preise" }, "url": "https://openai.com/index/previewing-gpt-5-6-sol/", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI GPT-5.6 preview system card", "de": "OpenAI GPT-5.6-Preview-Systemkarte" }, "url": "https://deploymentsafety.openai.com/gpt-5-6-preview", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/pricing", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/gpt-5-6-luna" }, { "slug": "gpt-5-6-sol", "name": "GPT-5.6 Sol", "provider": "OpenAI", "family": "GPT-5.6", "summary": { "en": "OpenAI's July 2026 GPT-5.6 flagship, superseded as the current OpenAI flagship by GPT-6 Astra on 3 Sep 2026. Promo API list still $4 / $20 per 1M tokens (short context) through 21 Nov 2026.", "de": "OpenAIs GPT-5.6-Flagship vom Juli 2026, als aktuelles OpenAI-Flagship am 3. Sep. 2026 von GPT-6 Astra abgelöst. Promo-API-Liste weiter $4 / $20 pro 1M Tokens (kurzer Kontext) bis 21. Nov. 2026." }, "role": { "en": "July 2026 GPT-5.6 Sol line. GPT-6 Sol (22 Sep 2026) is the current Sol tier. This promo list is unchanged. Prefer sourced Vals and AA ledger rows.", "de": "GPT-5.6-Sol-Linie vom Juli 2026. GPT-6 Sol (22. Sep. 2026) ist die aktuelle Sol-Stufe. Diese Promo-Liste ist unverändert. Bevorzuge belegte Vals- und AA-Ledger-Zeilen." }, "releaseDate": "2026-07-09", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 4, "outputPriceUSD": 20, "pricingNote": { "en": "OpenAI API promotional list through 21 Nov 2026: $4 input / $20 output per 1M tokens on short context (cached input $0.40). Long-context list is $8 / $30. Previously $5 / $30 at GA. Cache writes billed at uncached input; Fast mode is 2×.", "de": "OpenAI-API-Promo-Liste bis 21. Nov. 2026: $4 Input / $20 Output pro 1M Tokens bei kurzem Kontext (Cached Input $0,40). Long-Context-Liste $8 / $30. Zuvor $5 / $30 bei GA. Cache Writes zum Uncached-Input-Preis; Fast-Mode ist 2×." }, "pricingChecked": "2026-09-01", "status": "flagship", "sources": [ { "label": { "en": "OpenAI GPT-5.6 launch", "de": "OpenAI-GPT-5.6-Launch" }, "url": "https://openai.com/index/gpt-5-6/", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI GPT-5.6 Sol preview", "de": "OpenAI GPT-5.6-Sol-Preview" }, "url": "https://openai.com/index/previewing-gpt-5-6-sol/", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI GPT-5.6 preview system card", "de": "OpenAI GPT-5.6-Preview-Systemkarte" }, "url": "https://deploymentsafety.openai.com/gpt-5-6-preview", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://platform.openai.com/docs/pricing", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/gpt-5-6-sol" }, { "slug": "gpt-5-6-terra", "name": "GPT-5.6 Terra", "provider": "OpenAI", "family": "GPT-5.6", "summary": { "en": "OpenAI's generally available balanced GPT-5.6 model, positioned below Sol. API list for short context, checked 23 Sep 2026, is $2 / $12 per 1M tokens.", "de": "OpenAIs allgemein verfügbares, ausgewogenes GPT-5.6-Modell, unter Sol positioniert. Die API-Liste für kurzen Kontext, geprüft am 23. Sep. 2026, ist $2 / $12 pro 1M Tokens." }, "role": { "en": "Balanced GPT-5.6 tier for daily coding, research, and agent workflows when Sol's higher price is unnecessary.", "de": "Ausgewogene GPT-5.6-Stufe für tägliche Coding-, Research- und Agent-Workflows, wenn Sols höherer Preis unnötig ist." }, "releaseDate": "2026-07-09", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 12, "pricingNote": { "en": "OpenAI API standard list checked 23 Sep 2026: $2 input / $12 output per 1M tokens on short context. Cached input $0.20; cache writes $2.50. Context limit is still unpublished.", "de": "OpenAI-API-Standardliste geprüft am 23. Sep. 2026: $2 Input / $12 Output pro 1M Tokens bei kurzem Kontext. Cached Input $0,20; Cache-Writes $2,50. Das Kontextlimit ist weiter unveröffentlicht." }, "pricingChecked": "2026-09-23", "status": "solid", "sources": [ { "label": { "en": "OpenAI GPT-5.6 launch", "de": "OpenAI-GPT-5.6-Launch" }, "url": "https://openai.com/index/gpt-5-6/", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI GPT-5.6 preview availability and pricing", "de": "OpenAI GPT-5.6-Preview-Verfügbarkeit und Preise" }, "url": "https://openai.com/index/previewing-gpt-5-6-sol/", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI GPT-5.6 preview system card", "de": "OpenAI GPT-5.6-Preview-Systemkarte" }, "url": "https://deploymentsafety.openai.com/gpt-5-6-preview", "checkedAt": "2026-07-19" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/pricing", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/gpt-5-6-terra" }, { "slug": "gpt-5-codex", "name": "GPT-5 Codex (high)", "provider": "OpenAI", "family": "GPT-5 Codex", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$10 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$10 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-09-23", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 10, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/gpt-5-codex" }, { "slug": "gpt-5-low", "name": "GPT-5 (low)", "provider": "OpenAI", "family": "GPT-5 (low)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$10 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$10 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-08-07", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 10, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/gpt-5-low" }, { "slug": "gpt-5-medium", "name": "GPT-5 (medium)", "provider": "OpenAI", "family": "GPT-5", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$10 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$10 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-08-07", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 10, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/gpt-5-medium" }, { "slug": "gpt-5", "name": "GPT-5 (high)", "provider": "OpenAI", "family": "GPT-5 (high)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$10 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$10 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-08-07", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 10, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/gpt-5" }, { "slug": "gpt-6-1-sol", "name": "GPT-6.1 Sol", "provider": "OpenAI", "family": "GPT-6", "summary": { "en": "OpenAI's 29 Sep 2026 Sol line. API list $2 / $10 per 1M tokens on short context, 1.05M context, 128K output, knowledge cutoff 30 Apr 2026. Cache reads are 5% of input. No AA or Vals ledger row on this page.", "de": "OpenAIs Sol-Linie vom 29. Sep. 2026. API-Liste $2 / $10 pro 1M Tokens bei kurzem Kontext, 1,05M-Kontext, 128K-Output, Knowledge-Cutoff 30. Apr. 2026. Cache-Reads liegen bei 5 % des Inputs. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite." }, "role": { "en": "Current OpenAI Sol line for complex coding, computer use, and professional work when Astra's list is too expensive.", "de": "Aktuelle OpenAI-Sol-Linie für komplexes Coding, Computer-Nutzung und professionelle Arbeit, wenn Astras Liste zu teuer ist." }, "releaseDate": "2026-09-29", "knowledgeCutoff": "2026-04-30", "contextWindow": 1050000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 10, "pricingNote": { "en": "OpenAI API standard list checked 29 Sep 2026: $2 input / $10 output per 1M tokens on short context. Cached input $0.10; cache writes $2.50. Prompts over 272K input tokens are 2× input and cache and 1.5× output for the full request. Batch and Flex are half of Standard; Fast mode is 2×. Regional processing adds 10%. Fast mode is unavailable with EU data residency.", "de": "OpenAI-API-Standardliste geprüft am 29. Sep. 2026: $2 Input / $10 Output pro 1M Tokens bei kurzem Kontext. Cached Input $0,10; Cache-Writes $2,50. Prompts über 272K Input-Tokens sind 2× Input und Cache und 1,5× Output für den ganzen Request. Batch und Flex sind die Hälfte von Standard; Fast-Mode ist 2×. Regionale Verarbeitung kostet 10 % extra. Fast-Mode ist mit EU-Datenresidenz nicht verfügbar." }, "pricingChecked": "2026-09-29", "status": "flagship", "sources": [ { "label": { "en": "OpenAI API model page", "de": "OpenAI-API-Modellseite" }, "url": "https://developers.openai.com/api/docs/models/gpt-6.1-sol", "checkedAt": "2026-09-29" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/pricing", "checkedAt": "2026-09-29" }, { "label": { "en": "OpenAI deployment safety addendum", "de": "OpenAI-Deployment-Safety-Addendum" }, "url": "https://deploymentsafety.openai.com/gpt-6-1-sol/introduction", "checkedAt": "2026-09-29" } ], "html": "https://verdictpal.com/models/gpt-6-1-sol" }, { "slug": "gpt-6-astra", "name": "GPT-6 Astra", "provider": "OpenAI", "family": "GPT-6", "summary": { "en": "OpenAI's 3 Sep 2026 flagship. API list $10 / $50 per 1M tokens, 1.05M context, 128K output, knowledge cutoff 30 Apr 2026. ChatGPT ships it as GPT-6 Pro on eligible plans.", "de": "OpenAIs Flaggschiff vom 3. Sep. 2026. API-Liste $10 / $50 pro 1M Tokens, 1,05M-Kontext, 128K-Output, Knowledge-Cutoff 30. Apr. 2026. ChatGPT liefert es auf berechtigten Plänen als GPT-6 Pro." }, "role": { "en": "Current OpenAI flagship for long-horizon coding, computer use, research, and document work.", "de": "Aktuelles OpenAI-Flagship für Long-Horizon-Coding, Computer-Nutzung, Research und Dokumentarbeit." }, "releaseDate": "2026-09-03", "knowledgeCutoff": "2026-04-30", "contextWindow": 1050000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 10, "outputPriceUSD": 50, "pricingNote": { "en": "OpenAI API list $10 input / $50 output per 1M tokens. Cached input $1.00; cache writes $12.50. Prompts over 272K input tokens are 2× input/cache and 1.5× output for the full request. Batch and Flex are 50% of Standard; Fast mode is 2×. reasoning.effort supports low, medium, high, xhigh, and max. Model ID gpt-6-astra.", "de": "OpenAI-API-Liste $10 Input / $50 Output pro 1M Tokens. Cached Input $1,00; Cache-Writes $12,50. Prompts über 272K Input-Tokens sind 2× Input/Cache und 1,5× Output für den ganzen Request. Batch und Flex sind 50 % von Standard; Fast-Mode ist 2×. reasoning.effort unterstützt low, medium, high, xhigh und max. Modell-ID gpt-6-astra." }, "pricingChecked": "2026-09-18", "status": "flagship", "sources": [ { "label": { "en": "OpenAI GPT-6 Astra announcement", "de": "OpenAI-GPT-6-Astra-Ankündigung" }, "url": "https://openai.com/index/gpt-6-astra/", "checkedAt": "2026-09-18" }, { "label": { "en": "OpenAI API model page", "de": "OpenAI-API-Modellseite" }, "url": "https://developers.openai.com/api/docs/models/gpt-6-astra", "checkedAt": "2026-09-18" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/pricing", "checkedAt": "2026-09-18" }, { "label": { "en": "Artificial Analysis GPT-6 Astra", "de": "Artificial-Analysis-GPT-6-Astra" }, "url": "https://artificialanalysis.ai/models/gpt-6-astra", "checkedAt": "2026-09-10" }, { "label": { "en": "Vals model page", "de": "Vals-Modellseite" }, "url": "https://www.vals.ai/models/openai_gpt-6-astra", "checkedAt": "2026-09-05" } ], "html": "https://verdictpal.com/models/gpt-6-astra" }, { "slug": "gpt-6-luna", "name": "GPT-6 Luna", "provider": "OpenAI", "family": "GPT-6", "summary": { "en": "OpenAI's 22 Sep 2026 Luna line for high-volume work. API list $0.10 / $0.50 per 1M tokens on short context, 1.05M context, 128K output, knowledge cutoff 18 May 2026. No AA or Vals ledger row on this page as of 23 Sep 2026.", "de": "OpenAIs Luna-Linie vom 22. Sep. 2026 für High-Volume-Arbeit. API-Liste $0,10 / $0,50 pro 1M Tokens bei kurzem Kontext, 1,05M-Kontext, 128K-Output, Knowledge-Cutoff 18. Mai 2026. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite Stand 23. Sep. 2026." }, "role": { "en": "Current OpenAI Luna line for high-volume extraction, routing, and everyday agent calls.", "de": "Aktuelle OpenAI-Luna-Linie für High-Volume-Extraktion, Routing und alltägliche Agentenaufrufe." }, "releaseDate": "2026-09-22", "knowledgeCutoff": "2026-05-18", "contextWindow": 1050000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 0.1, "outputPriceUSD": 0.5, "pricingNote": { "en": "OpenAI API standard list checked 29 Sep 2026: $0.10 input / $0.50 output per 1M tokens on short context. Cached input $0.01; cache writes $0.125. Prompts over 272K input tokens are 2× input and cache and 1.5× output for the full request. Batch and Flex are half of Standard; Fast mode is 2×. EU data residency covers Standard, Flex, and Batch.", "de": "OpenAI-API-Standardliste geprüft am 29. Sep. 2026: $0,10 Input / $0,50 Output pro 1M Tokens bei kurzem Kontext. Cached Input $0,01; Cache-Writes $0,125. Prompts über 272K Input-Tokens sind 2× Input und Cache und 1,5× Output für den ganzen Request. Batch und Flex sind die Hälfte von Standard; Fast-Mode ist 2×. EU-Datenresidenz gilt für Standard, Flex und Batch." }, "pricingChecked": "2026-09-29", "status": "flagship", "sources": [ { "label": { "en": "OpenAI API model page", "de": "OpenAI-API-Modellseite" }, "url": "https://developers.openai.com/api/docs/models/gpt-6-luna", "checkedAt": "2026-09-29" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/pricing", "checkedAt": "2026-09-29" }, { "label": { "en": "OpenAI developer announcement", "de": "OpenAI-Entwicklerankündigung" }, "url": "https://community.openai.com/t/announcing-gpt-6-sol-and-gpt-6-luna-in-the-api-codex-and-chatgpt/1399925", "checkedAt": "2026-09-23" }, { "label": { "en": "Vercel AI Gateway changelog", "de": "Vercel-AI-Gateway-Changelog" }, "url": "https://vercel.com/changelog/gpt-6-sol-and-luna-now-available-on-ai-gateway", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/gpt-6-luna" }, { "slug": "gpt-6-sol", "name": "GPT-6 Sol", "provider": "OpenAI", "family": "GPT-6", "summary": { "en": "OpenAI's 22 Sep 2026 Sol line. API list $2 / $10 per 1M tokens on short context, 1.05M context, 128K output, knowledge cutoff 20 Apr 2026. No AA or Vals ledger row on this page as of 23 Sep 2026.", "de": "OpenAIs Sol-Linie vom 22. Sep. 2026. API-Liste $2 / $10 pro 1M Tokens bei kurzem Kontext, 1,05M-Kontext, 128K-Output, Knowledge-Cutoff 20. Apr. 2026. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite Stand 23. Sep. 2026." }, "role": { "en": "Current OpenAI Sol line for complex coding and agent work when Astra's list is too expensive.", "de": "Aktuelle OpenAI-Sol-Linie für komplexes Coding und Agentenarbeit, wenn Astras Liste zu teuer ist." }, "releaseDate": "2026-09-22", "knowledgeCutoff": "2026-04-20", "contextWindow": 1050000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 10, "pricingNote": { "en": "OpenAI API standard list checked 29 Sep 2026: $2 input / $10 output per 1M tokens on short context. Cached input $0.20; cache writes $2.50. Prompts over 272K input tokens are 2× input and cache and 1.5× output for the full request. Batch and Flex are half of Standard; Fast mode is 2×. EU data residency covers Standard, Flex, and Batch.", "de": "OpenAI-API-Standardliste geprüft am 29. Sep. 2026: $2 Input / $10 Output pro 1M Tokens bei kurzem Kontext. Cached Input $0,20; Cache-Writes $2,50. Prompts über 272K Input-Tokens sind 2× Input und Cache und 1,5× Output für den ganzen Request. Batch und Flex sind die Hälfte von Standard; Fast-Mode ist 2×. EU-Datenresidenz gilt für Standard, Flex und Batch." }, "pricingChecked": "2026-09-29", "status": "flagship", "sources": [ { "label": { "en": "OpenAI API model page", "de": "OpenAI-API-Modellseite" }, "url": "https://developers.openai.com/api/docs/models/gpt-6-sol", "checkedAt": "2026-09-29" }, { "label": { "en": "OpenAI API pricing", "de": "OpenAI-API-Preise" }, "url": "https://developers.openai.com/api/docs/pricing", "checkedAt": "2026-09-29" }, { "label": { "en": "OpenAI developer announcement", "de": "OpenAI-Entwicklerankündigung" }, "url": "https://community.openai.com/t/announcing-gpt-6-sol-and-gpt-6-luna-in-the-api-codex-and-chatgpt/1399925", "checkedAt": "2026-09-23" }, { "label": { "en": "Vercel AI Gateway changelog", "de": "Vercel-AI-Gateway-Changelog" }, "url": "https://vercel.com/changelog/gpt-6-sol-and-luna-now-available-on-ai-gateway", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/gpt-6-sol" }, { "slug": "granite-4-1-30b", "name": "Granite 4.1 30B", "provider": "Other", "family": "Granite 4.1", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-29", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/granite-4-1-30b" }, { "slug": "granite-4-2-30b", "name": "Granite 4.2 30B", "provider": "Other", "family": "Granite 4.2", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.16/$0.65 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.16/$0.65 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-25", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.16, "outputPriceUSD": 0.65, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/granite-4-2-30b" }, { "slug": "granite-4-2-3b", "name": "Granite 4.2 3B", "provider": "Other", "family": "Granite 4.2", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.03/$0.12 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.03/$0.12 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-25", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.03, "outputPriceUSD": 0.12, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/granite-4-2-3b" }, { "slug": "granite-4-2-8b", "name": "Granite 4.2 8B", "provider": "Other", "family": "Granite 4.2", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.06/$0.25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.06/$0.25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-25", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.06, "outputPriceUSD": 0.25, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/granite-4-2-8b" }, { "slug": "grok-4-20-0309", "name": "Grok 4.20 0309 (Reasoning)", "provider": "xAI", "family": "Grok 4.20", "summary": { "en": "Listed on Artificial Analysis with API pricing $2/$6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $2/$6 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-10", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/grok-4-20-0309" }, { "slug": "grok-4-20-non-reasoning", "name": "Grok 4.20 0309 v2 (Non-reasoning)", "provider": "xAI", "family": "Grok 4.20", "summary": { "en": "Listed on Artificial Analysis with API pricing $2/$6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $2/$6 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-07", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/grok-4-20-non-reasoning" }, { "slug": "grok-4-20", "name": "Grok 4.20 Reasoning", "provider": "xAI", "family": "Grok 4", "summary": { "en": "xAI's March 2026 reasoning Grok. Reasoning-first variant of the Grok 4 family.", "de": "xAIs Reasoning-Grok März 2026. Reasoning-first-Variante der Grok-4-Familie." }, "role": { "en": "xAI reasoning route when Grok 4.3 is too expensive or too general.", "de": "xAI-Reasoning-Route, wenn Grok 4.3 zu teuer oder zu allgemein ist." }, "releaseDate": "2026-03-05", "knowledgeCutoff": "2026-01-01", "contextWindow": 256000, "maxOutput": 65536, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "xAI API pricing varies by tier — check docs.x.ai.", "de": "xAI-API-Preise variieren nach Tier — docs.x.ai prüfen." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "xAI docs", "de": "xAI-Docs" }, "url": "https://docs.x.ai/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/grok-4-20" }, { "slug": "grok-4-3-low", "name": "Grok 4.3 (low)", "provider": "xAI", "family": "Grok 4.3", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-30", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 2.5, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/grok-4-3-low" }, { "slug": "grok-4-3-medium", "name": "Grok 4.3 (medium)", "provider": "xAI", "family": "Grok 4.3", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-30", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 2.5, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/grok-4-3-medium" }, { "slug": "grok-4-3-non-reasoning", "name": "Grok 4.3 (Non-reasoning)", "provider": "xAI", "family": "Grok 4.3", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.25/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.25/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-30", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 1.25, "outputPriceUSD": 2.5, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/grok-4-3-non-reasoning" }, { "slug": "grok-4-3", "name": "Grok 4.3", "provider": "xAI", "family": "Grok", "summary": { "en": "Not the smartest model on the board, but the only one with a 2M context and a live social graph.", "de": "Nicht das klügste Modell der Bestenliste, aber das einzige mit 2M-Kontext und einem Live-Sozialgraphen." }, "role": { "en": "Default model for real-time social signals, very long context (>1M), and freshness-sensitive research where minutes matter.", "de": "Default-Modell für Echtzeit-Soziale-Signale, sehr langen Kontext (>1M) und aktualitätskritische Recherche, in der Minuten zählen." }, "releaseDate": "2026-04-30", "knowledgeCutoff": "2026-03-15", "contextWindow": 2000000, "maxOutput": 128000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 5, "outputPriceUSD": 25, "pricingNote": { "en": "Real-time X search add-on is billed separately. 2M context is the only true >1M public context as of June 2026.", "de": "Echtzeit-X-Such-Add-on separat abgerechnet. 2M-Kontext ist der einzige echte öffentliche >1M-Kontext Stand Juni 2026." }, "pricingChecked": "2026-06-01", "status": "solid", "sources": [ { "label": { "en": "xAI pricing", "de": "xAI-Preise" }, "url": "https://x.ai/pricing", "checkedAt": "2026-06-01" }, { "label": { "en": "Grok 4.3 release", "de": "Grok-4.3-Release" }, "url": "https://x.ai/news/grok-4-3", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/grok-4-3" }, { "slug": "grok-4-5", "name": "Grok 4.5", "provider": "xAI", "family": "Grok", "summary": { "en": "xAI's July 2026 frontier release. API list price $2 / $6 per 1M tokens.", "de": "xAIs Frontier-Release Juli 2026. API-Listenpreis $2 / $6 pro 1M Tokens." }, "role": { "en": "Frontier xAI model for long-context and freshness-sensitive work. Prefer sourced ledger rows over marketing claims; AA scores sync from the public snapshot.", "de": "Frontier-xAI-Modell für langen Kontext und aktualitätskritische Arbeit. Bevorzuge belegte Ledger-Zeilen gegenüber Marketing; AA-Scores kommen aus dem öffentlichen Snapshot." }, "releaseDate": "2026-07-08", "knowledgeCutoff": "open", "contextWindow": 2000000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingNote": { "en": "AA list pricing for the high-effort Grok 4.5 endpoint as of 2026-07-09: $2 input / $6 output per 1M tokens.", "de": "AA-Listenpreise für den High-Effort-Grok-4.5-Endpoint Stand 2026-07-09: $2 Input / $6 Output pro 1M Tokens." }, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" }, { "label": { "en": "xAI Grok", "de": "xAI Grok" }, "url": "https://x.ai/grok", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/grok-4-5" }, { "slug": "grok-4-6", "name": "Grok 4.6", "provider": "xAI", "family": "Grok", "summary": { "en": "xAI's August 2026 flagship. API list $2 / $6 per 1M tokens; AA lists a 500k context window.", "de": "xAIs Flaggschiff August 2026. API-Liste $2 / $6 pro 1M Tokens; AA listet 500k Kontext." }, "role": { "en": "Previous xAI flagship. Grok 4.7 (21 Sep 2026) is the current line at the same starting list. Prefer sourced ledger rows over launch claims; AA scores sync from the public snapshot.", "de": "Vorheriges xAI-Flagship. Grok 4.7 (21. Sep. 2026) ist die aktuelle Linie zur selben Startliste. Bevorzuge belegte Ledger-Zeilen gegenüber Launch-Claims; AA-Scores kommen aus dem öffentlichen Snapshot." }, "releaseDate": "2026-08-12", "knowledgeCutoff": "open", "contextWindow": 500000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingNote": { "en": "AA list pricing for Grok 4.6 (high): $2 input / $6 output per 1M tokens. Cache-hit input is listed at $0.50 per 1M. The live AA model page on 5 Sep 2026 lists a 500k context window.", "de": "AA-Listenpreise für Grok 4.6 (high): $2 Input / $6 Output pro 1M Tokens. Cache-Hit-Input steht mit $0,50 pro 1M. Die live AA-Modellseite listet am 5. Sep. 2026 ein 500k-Kontextfenster." }, "pricingChecked": "2026-09-05", "status": "flagship", "sources": [ { "label": { "en": "xAI Grok 4.6 announcement", "de": "xAI-Grok-4.6-Ankündigung" }, "url": "https://x.ai/news/grok-4-6", "checkedAt": "2026-08-12" }, { "label": { "en": "Artificial Analysis Grok 4.6", "de": "Artificial-Analysis-Grok-4.6" }, "url": "https://artificialanalysis.ai/models/grok-4-6", "checkedAt": "2026-09-10" }, { "label": { "en": "xAI Grok", "de": "xAI Grok" }, "url": "https://x.ai/grok", "checkedAt": "2026-08-12" } ], "html": "https://verdictpal.com/models/grok-4-6" }, { "slug": "grok-4-7", "name": "Grok 4.7", "provider": "xAI", "family": "Grok", "summary": { "en": "xAI's 21 Sep 2026 Grok model for coding and knowledge work. API list starts at $2 / $6 per 1M tokens, 500k context. Text and image in, text out. No AA or Vals ledger row on this page as of 23 Sep 2026.", "de": "xAIs Grok-Modell vom 21. Sep. 2026 für Coding und Wissensarbeit. API-Liste beginnt bei $2 / $6 pro 1M Tokens, 500k-Kontext. Text und Bild rein, Text raus. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite Stand 23. Sep. 2026." }, "role": { "en": "Current xAI line for agentic coding and knowledge work. Grok 4.6 stays on the atlas at the same starting list.", "de": "Aktuelle xAI-Linie für agentisches Coding und Wissensarbeit. Grok 4.6 bleibt im Atlas zur selben Startliste." }, "releaseDate": "2026-09-21", "knowledgeCutoff": "unpublished", "contextWindow": 500000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingNote": { "en": "xAI model docs checked 23 Sep 2026: under 200k prompt tokens, $2 input / $0.50 cached input / $6 output per 1M tokens. At or above 200k prompt tokens the whole request is $4 / $1 / $12. Default reasoning effort is high. The launch post also describes a fast variant at twice the output speed and twice the price; the docs page lists one model.", "de": "xAI-Modelldoku geprüft am 23. Sep. 2026: unter 200k Prompt-Tokens $2 Input / $0,50 Cached Input / $6 Output pro 1M Tokens. Ab 200k Prompt-Tokens gilt für den ganzen Request $4 / $1 / $12. Standard-Reasoning-Effort ist high. Der Launch-Post beschreibt zusätzlich eine Fast-Variante mit doppelter Output-Geschwindigkeit zum doppelten Preis; die Doku listet ein Modell." }, "pricingChecked": "2026-09-23", "status": "flagship", "sources": [ { "label": { "en": "xAI Grok 4.7 announcement", "de": "xAI-Ankündigung Grok 4.7" }, "url": "https://x.ai/news/grok-4-7", "checkedAt": "2026-09-23" }, { "label": { "en": "xAI Grok 4.7 model docs", "de": "xAI-Modelldoku Grok 4.7" }, "url": "https://docs.x.ai/docs/models/grok-4.7", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/grok-4-7" }, { "slug": "grok-4", "name": "Grok 4", "provider": "xAI", "family": "Grok 4", "summary": { "en": "Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $3/$15 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-07-10", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/grok-4" }, { "slug": "grok-build-0-1-06-16", "name": "Grok Build 0.1 0616", "provider": "xAI", "family": "Grok Build", "summary": { "en": "Listed on Artificial Analysis with API pricing $1/$2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1/$2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-06-16", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1, "outputPriceUSD": 2, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/grok-build-0-1-06-16" }, { "slug": "hy3-non-reasoning", "name": "Hy3-preview (Non-reasoning)", "provider": "Other", "family": "Hy3-preview (Non-reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.065/$0.235 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.065/$0.235 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.065, "outputPriceUSD": 0.235, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/hy3-non-reasoning" }, { "slug": "hy3-preview", "name": "Hy3-preview (Reasoning)", "provider": "Other", "family": "Hy3-preview (Reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.065/$0.235 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.065/$0.235 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.065, "outputPriceUSD": 0.235, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/hy3-preview" }, { "slug": "hy3", "name": "Hy3", "provider": "Other", "family": "Other", "summary": { "en": "Tencent Hunyuan Hy3. Superseded as the current Hunyuan preview by Hy4 preview (28 Aug 2026, 770B / 49B, $0.834 / $2.501).", "de": "Tencent Hunyuan Hy3. Als aktuelle Hunyuan-Preview von Hy4 preview abgelöst (28. Aug. 2026, 770B / 49B, $0,834 / $2,501)." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-06", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/hy3" }, { "slug": "hy4-preview", "name": "Hy4 preview", "provider": "Other", "family": "Hy4", "summary": { "en": "Tencent Hunyuan's 28 Aug 2026 open-weight preview: 770B total / 49B active MoE, context over 1M tokens. API list $0.834 / $2.501 per 1M tokens; cache hits $0.042. Not yet on the 1 Sep AA API snapshot.", "de": "Tencent Hunyuans Open-Weight-Preview vom 28. Aug. 2026: 770B gesamt / 49B aktiv MoE, Kontext über 1M Tokens. API-Liste $0,834 / $2,501 pro 1M Tokens; Cache-Hits $0,042. Steht im AA-API-Snapshot vom 1. Sep. noch nicht." }, "role": { "en": "Open-weight Hunyuan preview for coding, office, and research agents. Prefer the Tencent launch note over unsourced leaderboard screenshots; AA rows will land when the snapshot lists the model.", "de": "Open-Weight-Hunyuan-Preview für Coding-, Office- und Research-Agents. Bevorzuge die Tencent-Launch-Notiz gegenüber unbelegten Leaderboard-Screenshots; AA-Zeilen kommen, wenn der Snapshot das Modell listet." }, "releaseDate": "2026-08-28", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.834, "outputPriceUSD": 2.501, "pricingNote": { "en": "Tencent launch list: USD 0.834 input / 2.501 output per 1M tokens; cache hits USD 0.042. Hosted via Tencent Cloud TokenHub and OpenRouter. WorkBuddy and CodeBuddy had a two-week free window at launch. Internal Tencent blind eval (163 experts, 203 engineering tasks) is vendor-claimed, not a ledger row.", "de": "Tencent-Launch-Liste: USD 0,834 Input / 2,501 Output pro 1M Tokens; Cache-Hits USD 0,042. Gehostet über Tencent Cloud TokenHub und OpenRouter. WorkBuddy und CodeBuddy hatten ein zweiwöchiges Free-Fenster beim Launch. Interne Tencent-Blind-Eval (163 Expert:innen, 203 Engineering-Aufgaben) ist Vendor-Claim, keine Ledger-Zeile." }, "pricingChecked": "2026-09-05", "status": "solid", "sources": [ { "label": { "en": "Tencent Hy4 preview announcement", "de": "Tencent-Hy4-preview-Ankündigung" }, "url": "https://www.tencent.com/tencent-releases-and-open-sources-tencent-hy4-preview/", "checkedAt": "2026-09-05" } ], "html": "https://verdictpal.com/models/hy4-preview" }, { "slug": "hypernova-60b-2605", "name": "HyperNova 60B 2605", "provider": "Multiverse", "family": "HyperNova", "summary": { "en": "Multiverse Computing's May 2026 open-source 60B update, evaluated by Artificial Analysis on 11 Jun. The release emphasizes coding, native tool calling, OpenAI-style function schemas, structured outputs, and configurable reasoning effort.", "de": "Multiverse Computings Open-Source-60B-Update Mai 2026, am 11. Juni von Artificial Analysis evaluiert. Der Release betont Coding, natives Tool-Calling, OpenAI-artige Function-Schemas, strukturierte Outputs und konfigurierbaren Reasoning-Aufwand." }, "role": { "en": "Compressed open model for coding agents and tool-augmented workflows when you want Apache-licensed weights instead of a closed API.", "de": "Komprimiertes Open-Modell für Coding-Agents und tool-gestützte Workflows, wenn Apache-lizenzierte Weights wichtiger sind als eine geschlossene API." }, "releaseDate": "2026-05-06", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Open weights on Hugging Face; no canonical API list price captured yet.", "de": "Open Weights auf Hugging Face; noch kein kanonischer API-Listenpreis erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Multiverse announcement", "de": "Multiverse-Ankündigung" }, "url": "https://multiversecomputing.com/resources/introducing-hypernova-60b-2605", "checkedAt": "2026-06-14" }, { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/hypernova-60b-2605" }, { "slug": "inkling-small", "name": "Inkling Small", "provider": "Other", "family": "Inkling Small", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$1.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$1.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-30", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/inkling-small" }, { "slug": "inkling", "name": "Inkling (xhigh)", "provider": "Other", "family": "Inkling (xhigh)", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.87/$4.68 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.87/$4.68 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-15", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.87, "outputPriceUSD": 4.68, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/inkling" }, { "slug": "jev", "name": "Jev", "provider": "TypeSafe", "family": "Jev", "summary": { "en": "TypeSafe's 15 Sep 2026 System One model. Text-only state in; typed Noul, Choice, and Score decisions out — no generated prose. 64K request budget (32K for state plus the longest question). List $0.042 per 1M input tokens; output free.", "de": "TypeSafes System-One-Modell vom 15. Sep. 2026. Nur Text-State rein; typisierte Noul-, Choice- und Score-Entscheidungen raus — keine generierte Prosa. 64K-Request-Budget (32K für State plus die längste Frage). Liste $0,042 pro 1M Input-Tokens; Output kostenlos." }, "role": { "en": "Current TypeSafe System One workhorse for software-native classify, route, and score calls — not a chat or coding model.", "de": "Aktuelles TypeSafe-System-One-Arbeitspferd für software-native Classify-, Route- und Score-Aufrufe — kein Chat- oder Coding-Modell." }, "releaseDate": "2026-09-15", "knowledgeCutoff": "unpublished", "contextWindow": 64000, "modalities": [ "text" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.042, "outputPriceUSD": 0, "pricingNote": { "en": "TypeSafe models docs (checked 19 Sep 2026): $0.042 per 1M input tokens; output tokens free. Early-access limits 250,000 tokens/second and 1,200 requests/minute, changing without notice. Aliases jev-latest and jev-preview currently resolve to jev-1.13.0. Vercel AI Gateway lists typesafe-ai/jev at the same input rate (changelog 16 Sep 2026). English is the primary training language.", "de": "TypeSafe-Models-Docs (geprüft 19. Sep. 2026): $0,042 pro 1M Input-Tokens; Output-Tokens kostenlos. Early-Access-Limits 250.000 Tokens/Sekunde und 1.200 Requests/Minute, ohne Vorankündigung. Aliase jev-latest und jev-preview zeigen derzeit auf jev-1.13.0. Vercel AI Gateway listet typesafe-ai/jev zum selben Input-Tarif (Changelog 16. Sep. 2026). Englisch ist die primäre Trainingsprache." }, "pricingChecked": "2026-09-19", "status": "solid", "sources": [ { "label": { "en": "TypeSafe System One and Jev announcement", "de": "TypeSafe-Ankündigung System One und Jev" }, "url": "https://typesafe.ai/blog/introducing-system-one-models-and-jev", "checkedAt": "2026-09-19" }, { "label": { "en": "TypeSafe models docs", "de": "TypeSafe-Models-Docs" }, "url": "https://docs.typesafe.ai/models", "checkedAt": "2026-09-19" }, { "label": { "en": "TypeSafe API reference", "de": "TypeSafe-API-Referenz" }, "url": "https://docs.typesafe.ai/api", "checkedAt": "2026-09-19" }, { "label": { "en": "TypeSafe Jev 1.13 jaggedness", "de": "TypeSafe Jev-1.13-Jaggedness" }, "url": "https://docs.typesafe.ai/model-jaggedness/jev-1.13", "checkedAt": "2026-09-19" }, { "label": { "en": "TypeSafe privacy policy", "de": "TypeSafe-Datenschutz" }, "url": "https://typesafe.ai/privacy", "checkedAt": "2026-09-19" }, { "label": { "en": "Vercel AI Gateway Jev", "de": "Vercel-AI-Gateway Jev" }, "url": "https://vercel.com/ai-gateway/models/jev", "checkedAt": "2026-09-19" }, { "label": { "en": "Vercel AI Gateway changelog", "de": "Vercel-AI-Gateway-Changelog" }, "url": "https://vercel.com/changelog/typesafe-ai-jev-now-available-on-ai-gateway", "checkedAt": "2026-09-19" } ], "html": "https://verdictpal.com/models/jev" }, { "slug": "jt-35b-flash", "name": "JT-35B-Flash", "provider": "Other", "family": "Other", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-14", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/jt-35b-flash" }, { "slug": "jt-4-1-flash-236b-a21b", "name": "JT-4.1 Flash 236B A21B", "provider": "Other", "family": "JT-4.1 Flash", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-09", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/jt-4-1-flash-236b-a21b" }, { "slug": "jt-mini", "name": "JT-MINI", "provider": "Other", "family": "Other", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-15", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/jt-mini" }, { "slug": "k-exaone-2-0-0803", "name": "K-EXAONE 2.0 0803", "provider": "Other", "family": "EXAONE", "summary": { "en": "LG AI Research K-EXAONE 2.0 (750B-class Korean sovereign model, 31 Jul 2026 weights). AA lists $0 / $0 per 1M tokens.", "de": "LG AI Research K-EXAONE 2.0 (koreanisches Sovereign-Modell der 750B-Klasse, Gewichte 31. Jul. 2026). AA führt $0 / $0 pro 1M Tokens." }, "role": { "en": "Open Korean foundation-model lane. Prefer sourced ledger rows over ministry launch copy.", "de": "Offene koreanische Foundation-Model-Spur. Bevorzuge belegte Ledger-Zeilen gegenüber Ministeriums-Launch-Text." }, "releaseDate": "2026-08-12", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingNote": { "en": "LG says Apache 2.0 weights on Hugging Face (31 Jul 2026). AA snapshot 2026-08-15 lists $0 / $0 per 1M tokens for the 0803 endpoint.", "de": "LG nennt Apache-2.0-Gewichte auf Hugging Face (31. Jul. 2026). Der AA-Snapshot vom 2026-08-15 führt $0 / $0 pro 1M Tokens für den 0803-Endpoint." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "LG AI Research K-EXAONE 2.0 news", "de": "LG-AI-Research-K-EXAONE-2.0-News" }, "url": "https://www.lgresearch.ai/news/view?seq=678", "checkedAt": "2026-08-15" }, { "label": { "en": "Artificial Analysis K-EXAONE 2.0 0803", "de": "Artificial-Analysis-K-EXAONE-2.0-0803" }, "url": "https://artificialanalysis.ai/models/k-exaone-2-0-0803", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/k-exaone-2-0-0803" }, { "slug": "k2-horizon-375b-a23b", "name": "K2 Horizon 375B A23B", "provider": "Other", "family": "K2 Horizon", "summary": { "en": "IFM (MBZUAI) 3 Sep 2026 flagship: 375B-A23B sparse MoE, 512K native context, Apache 2.0 weights. Official HF is IFM/K2-Horizon-375B-A23B. The 10 Sep AA snapshot lists API price as $0 / $0 — do not invent a hosted dollar rate.", "de": "IFM-(MBZUAI)-Flaggschiff vom 3. Sep. 2026: 375B-A23B Sparse-MoE, 512K nativer Kontext, Apache-2.0-Gewichte. Offizielles HF ist IFM/K2-Horizon-375B-A23B. Der AA-Snapshot vom 10. Sep. listet den API-Preis als $0 / $0 — keinen gehosteten Dollar-Satz erfinden." }, "role": { "en": "Open research fleet flagship for reproducing training and running a mid-size MoE locally or via IFM partners. Prefer the AA ledger row and the IFM launch note over vendor eval tables.", "de": "Open-Research-Flotten-Flaggschiff zum Nachbauen des Trainings und zum lokalen oder Partner-Lauf eines mittelgrossen MoE. Bevorzuge die AA-Ledger-Zeile und die IFM-Launch-Notiz gegenüber Vendor-Eval-Tabellen." }, "releaseDate": "2026-09-03", "knowledgeCutoff": "unpublished", "contextWindow": 524288, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingNote": { "en": "Weights and code are Apache 2.0. IFM lists day-zero serving via vLLM, SGLang, Ollama, and partners (Compass, Cerebras, AWS, Nebius). Artificial Analysis still records $0 / $0 per 1M tokens on the 10 Sep 2026 snapshot — no independent hosted list found.", "de": "Gewichte und Code stehen unter Apache 2.0. IFM nennt Day-Zero-Serving über vLLM, SGLang, Ollama und Partner (Compass, Cerebras, AWS, Nebius). Artificial Analysis verzeichnet im Snapshot vom 10. Sep. 2026 weiter $0 / $0 pro 1M Tokens — keine unabhängige Hosted-Liste gefunden." }, "pricingChecked": "2026-09-12", "status": "solid", "sources": [ { "label": { "en": "IFM K2 Horizon announcement", "de": "IFM-K2-Horizon-Ankündigung" }, "url": "https://ifm.ai/blog/k2/", "checkedAt": "2026-09-12" }, { "label": { "en": "Hugging Face K2-Horizon-375B-A23B", "de": "Hugging-Face-K2-Horizon-375B-A23B" }, "url": "https://huggingface.co/IFM/K2-Horizon-375B-A23B", "checkedAt": "2026-09-12" }, { "label": { "en": "Artificial Analysis K2 Horizon 375B A23B", "de": "Artificial-Analysis-K2-Horizon-375B-A23B" }, "url": "https://artificialanalysis.ai/models/k2-horizon-375b-a23b", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/k2-horizon-375b-a23b" }, { "slug": "kat-coder-pro-v2", "name": "KAT Coder Pro V2", "provider": "Other", "family": "KAT Coder", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$1.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$1.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-27", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/kat-coder-pro-v2" }, { "slug": "kimi-k2-5", "name": "Kimi K2.5", "provider": "Moonshot", "family": "Kimi", "summary": { "en": "Moonshot's January 2026 multimodal open-weights model. 262K context, $0.60/$3 per 1M tokens. Agent Swarm up to 100 sub-agents. Predecessor to K2.6.", "de": "Moonshots multimodales Open-Weights-Modell Januar 2026. 262K-Kontext, $0,60/$3 pro 1M Tokens. Agent Swarm mit bis zu 100 Sub-Agenten. Vorgänger von K2.6." }, "role": { "en": "Open-weights Moonshot pick when K2.6 isn't available or you need the K2.5 weight checkpoint.", "de": "Open-Weights-Moonshot-Wahl, wenn K2.6 nicht verfügbar ist oder der K2.5-Weight-Checkpoint benötigt wird." }, "releaseDate": "2026-01-27", "knowledgeCutoff": "2025-10-01", "contextWindow": 262144, "maxOutput": 65536, "modalities": [ "text", "image", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.6, "outputPriceUSD": 3, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Moonshot API pricing", "de": "Moonshot-API-Preise" }, "url": "https://platform.moonshot.ai/", "checkedAt": "2026-06-09" }, { "label": { "en": "Kimi K2.5 on Hugging Face", "de": "Kimi K2.5 auf Hugging Face" }, "url": "https://huggingface.co/moonshotai/Kimi-K2.5", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/kimi-k2-5" }, { "slug": "kimi-k2-6-non-reasoning", "name": "Kimi K2.6 (Non-reasoning)", "provider": "Other", "family": "Kimi K2.6", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.95/$4 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.95/$4 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-20", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.95, "outputPriceUSD": 4, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/kimi-k2-6-non-reasoning" }, { "slug": "kimi-k2-6", "name": "Kimi K2.6", "provider": "Moonshot", "family": "Kimi", "summary": { "en": "Moonshot's April 2026 open-weights flagship. Leads the open-source field on GPQA Diamond at 90.5%. 256K context, $0.60 / $2.50 per 1M tokens.", "de": "Moonshots Open-Weights-Flaggschiff April 2026. Führt das Open-Source-Feld bei GPQA Diamond mit 90,5 % an. 256K-Kontext, $0,60 / $2,50 pro 1M Tokens." }, "role": { "en": "Default open-weights model from a Chinese provider for English-language work. The strongest open-source model on GPQA Diamond (90.5%) and Vals (55.55%) as of June 2026.", "de": "Default-Open-Weights-Modell eines chinesischen Anbieters für englischsprachige Arbeit. Das stärkste Open-Source-Modell bei GPQA Diamond (90,5 %) und Vals (55,55 %) Stand Juni 2026." }, "releaseDate": "2026-04-20", "knowledgeCutoff": "2026-02-15", "contextWindow": 256000, "maxOutput": 65536, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.6, "outputPriceUSD": 2.5, "pricingNote": { "en": "Open weights under the Moonshot community license. API pricing is among the lowest for a frontier model.", "de": "Open Weights unter der Moonshot-Community-Lizenz. API-Preise gehören zu den niedrigsten für ein Frontier-Modell." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "Kimi K2.6 model card", "de": "Kimi-K2.6-Model-Karte" }, "url": "https://huggingface.co/moonshotai/Kimi-K2.6", "checkedAt": "2026-06-01" }, { "label": { "en": "Moonshot API pricing", "de": "Moonshot-API-Preise" }, "url": "https://platform.moonshot.cn/docs/pricing", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/kimi-k2-6" }, { "slug": "kimi-k2-7-code", "name": "Kimi K2.7 Code", "provider": "Moonshot", "family": "Kimi", "summary": { "en": "Moonshot's June 2026 coding-specialist Kimi release. Kimi's docs position `kimi-k2.7-code` as its most capable coding model, with a 256K context window, multimodal tool examples, mandatory thinking mode, and stronger long-horizon coding reliability than K2.6.", "de": "Moonshots coding-spezialisierter Kimi-Release Juni 2026. Kimi-Dokumente positionieren `kimi-k2.7-code` als fähigstes Coding-Modell, mit 256K-Kontextfenster, multimodalen Tool-Beispielen, verpflichtendem Thinking-Modus und stärkerer Long-Horizon-Coding-Zuverlässigkeit als K2.6." }, "role": { "en": "Budget coding agent model for repo-scale refactors, code review, MCP tool workflows, and long debugging loops where K2.6 was close but too token-hungry.", "de": "Budget-Coding-Agent-Modell für repo-weite Refactors, Code Review, MCP-Tool-Workflows und lange Debugging-Loops, bei denen K2.6 nah dran, aber zu tokenhungrig war." }, "releaseDate": "2026-06-12", "knowledgeCutoff": "open", "contextWindow": 262144, "maxOutput": 32768, "modalities": [ "text", "image", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.95, "outputPriceUSD": 4, "pricingNote": { "en": "Reported official API rate: $0.95/1M input, $4.00/1M output, $0.19/1M cache-hit input. Open weights are available on Hugging Face under Moonshot's Modified MIT license.", "de": "Berichteter offizieller API-Tarif: $0,95/1M Input, $4,00/1M Output, $0,19/1M Cache-Hit-Input. Open Weights sind auf Hugging Face unter Moonshots Modified-MIT-Lizenz verfügbar." }, "pricingChecked": "2026-06-14", "status": "solid", "sources": [ { "label": { "en": "Kimi K2.7 Code docs", "de": "Kimi-K2.7-Code-Dokumentation" }, "url": "https://platform.kimi.ai/docs/guide/kimi-k2-7-code-quickstart", "checkedAt": "2026-06-14" }, { "label": { "en": "Kimi Code", "de": "Kimi Code" }, "url": "https://www.kimi.com/code", "checkedAt": "2026-06-14" }, { "label": { "en": "Kimi K2.7 Code on Hugging Face", "de": "Kimi K2.7 Code auf Hugging Face" }, "url": "https://huggingface.co/moonshotai/Kimi-K2.7-Code", "checkedAt": "2026-06-14" }, { "label": { "en": "MarkTechPost release coverage", "de": "MarkTechPost-Release-Bericht" }, "url": "https://www.marktechpost.com/2026/06/12/moonshot-ai-releases-kimi-k2-7-code-a-coding-model-reporting-21-8-on-kimi-code-bench-v2-over-k2-6/", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/kimi-k2-7-code" }, { "slug": "kimi-k2-thinking", "name": "Kimi K2 Thinking", "provider": "Moonshot", "family": "Kimi K2", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.6/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.6/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-11-06", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.6, "outputPriceUSD": 2.5, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/kimi-k2-thinking" }, { "slug": "kimi-k3-low", "name": "Kimi K3 (low)", "provider": "Moonshot", "family": "Kimi K3", "summary": { "en": "Listed on Artificial Analysis with API pricing $3/$15 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $3/$15 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/kimi-k3-low" }, { "slug": "kimi-k3", "name": "Kimi K3", "provider": "Moonshot", "family": "Kimi K3", "summary": { "en": "Moonshot AI's 2.8T-parameter multimodal reasoning model for long-horizon coding and knowledge work, with a 1M-token context window, public API access, and open weights on Hugging Face.", "de": "Moonshot AIs multimodales Reasoning-Modell mit 2,8 Billionen Parametern für langlaufendes Coding und Wissensarbeit, mit 1 Mio. Token Kontext, öffentlichem API-Zugang und offenen Gewichten auf Hugging Face." }, "role": { "en": "Open-frontier option for large-repository coding, tool use, visual reasoning, and context-heavy research workflows — via API or self-hosted weights.", "de": "Open-Frontier-Option für Coding in grossen Repositories, Tool-Nutzung, visuelles Reasoning und kontextreiche Research-Workflows — per API oder selbst gehosteten Gewichten." }, "releaseDate": "2026-07-16", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "open-weights", "managed-cloud" ], "inputPriceUSD": 3, "outputPriceUSD": 15, "pricingNote": { "en": "Kimi API list price for cache-miss input and output. Cache-hit input is listed at $0.30 per 1M tokens. Amazon Bedrock standard global profile matches this list; the US profile is higher. Checked 23 Sep 2026.", "de": "Kimi-API-Listenpreis für Cache-Miss-Input und Output. Cache-Hit-Input ist mit $0,30 pro 1 Mio. Token gelistet. Das Amazon-Bedrock-Standardprofil global entspricht dieser Liste; das US-Profil liegt höher. Geprüft am 23. Sep. 2026." }, "pricingChecked": "2026-09-23", "status": "solid", "sources": [ { "label": { "en": "Kimi K3 launch and technical overview", "de": "Kimi-K3-Launch und technische Übersicht" }, "url": "https://www.kimi.com/blog/kimi-k3", "checkedAt": "2026-07-27" }, { "label": { "en": "Kimi K3 open weights on Hugging Face", "de": "Kimi-K3-Open-Weights auf Hugging Face" }, "url": "https://huggingface.co/moonshotai/Kimi-K3", "checkedAt": "2026-07-27" }, { "label": { "en": "Kimi K3 API pricing", "de": "Kimi-K3-API-Preise" }, "url": "https://platform.kimi.com/docs/pricing/chat-k3", "checkedAt": "2026-07-27" }, { "label": { "en": "Artificial Analysis model record", "de": "Artificial-Analysis-Modellnachweis" }, "url": "https://artificialanalysis.ai/models/kimi-k3", "checkedAt": "2026-07-27" }, { "label": { "en": "Amazon Bedrock Kimi K3", "de": "Amazon Bedrock Kimi K3" }, "url": "https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-moonshot-ai-kimi-k3.html", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/kimi-k3" }, { "slug": "lfm2-5-8b-a1b", "name": "LFM2.5-8B-A1B", "provider": "Liquid", "family": "LFM", "summary": { "en": "Liquid's compact LFM 2.5 model evaluated by Artificial Analysis on 8 Jun. Keep it in the efficiency lane until local benchmark and pricing rows are imported.", "de": "Liquids kompaktes LFM-2.5-Modell, am 8. Juni von Artificial Analysis evaluiert. Bis lokale Benchmark- und Pricing-Zeilen importiert sind, bleibt es in der Effizienz-Lane." }, "role": { "en": "Tiny efficiency model for routing, extraction, and on-device-adjacent experiments where footprint matters more than peak reasoning.", "de": "Kleines Effizienzmodell für Routing, Extraktion und On-Device-nahe Experimente, bei denen Footprint wichtiger ist als Spitzen-Reasoning." }, "releaseDate": "2026-06-08", "knowledgeCutoff": "open", "contextWindow": 128000, "maxOutput": 8192, "modalities": [ "text", "code" ], "access": [ "api", "on-device" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "No canonical public list price captured in the local ledger yet.", "de": "Noch kein kanonischer öffentlicher Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/lfm2-5-8b-a1b" }, { "slug": "ling-2-6-1t", "name": "Ling-2.6-1T", "provider": "Other", "family": "Other", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-23", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 2.5, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/ling-2-6-1t" }, { "slug": "ling-2-6-flash", "name": "Ling 2.6 Flash", "provider": "Other", "family": "Ling 2.6", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.1/$0.3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.1/$0.3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-21", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.1, "outputPriceUSD": 0.3, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/ling-2-6-flash" }, { "slug": "ling-3-0-flash", "name": "Ling 3.0 Flash", "provider": "Other", "family": "Ling 3.0", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.075/$0.22 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.075/$0.22 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-04", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.075, "outputPriceUSD": 0.22, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/ling-3-0-flash" }, { "slug": "ling-3-0-tiny", "name": "Ling 3.0 Tiny", "provider": "Other", "family": "Ling 3.0", "summary": { "en": "InclusionAI Ling 3.0 Tiny. AA lists $0 / $0 per 1M tokens.", "de": "InclusionAI Ling 3.0 Tiny. AA führt $0 / $0 pro 1M Tokens." }, "role": { "en": "Small Ling 3.0 sibling. Prefer sourced ledger rows; this is not a frontier face-off model.", "de": "Kleine Ling-3.0-Schwester. Bevorzuge belegte Ledger-Zeilen; das ist kein Frontier-Duell-Modell." }, "releaseDate": "2026-08-06", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-15", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis Ling 3.0 Tiny", "de": "Artificial-Analysis-Ling-3.0-Tiny" }, "url": "https://artificialanalysis.ai/models/ling-3-0-tiny", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/ling-3-0-tiny" }, { "slug": "llama-4-maverick", "name": "Llama 4 Maverick", "provider": "Meta", "family": "Llama 4", "summary": { "en": "Meta's reasoning-focused Llama 4 MoE. 1M context, 17B active / 400B total params, 128 experts. Open weights for multimodal assistants and coding.", "de": "Metas reasoning-fokussiertes Llama-4-MoE. 1M-Kontext, 17B aktiv / 400B gesamt, 128 Experten. Open Weights für multimodale Assistenten und Coding." }, "role": { "en": "Default Llama 4 pick for quality-first self-hosting when Scout's 10M context isn't required.", "de": "Standard-Llama-4-Wahl für qualitätsorientiertes Self-Hosting, wenn Scouts 10M-Kontext nicht nötig ist." }, "releaseDate": "2026-04-05", "knowledgeCutoff": "2024-08-01", "contextWindow": 1000000, "maxOutput": 16384, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "managed-cloud", "api" ], "inputPriceUSD": 0.17, "outputPriceUSD": 0.6, "pricingNote": { "en": "Typical hosted inference pricing; requires multi-GPU host for full weights. Community License free below 700M MAU.", "de": "Typische gehostete Inference-Preise; Multi-GPU-Host für volle Weights nötig. Community-Lizenz kostenlos unter 700M MAU." }, "pricingChecked": "2026-06-09", "status": "flagship", "sources": [ { "label": { "en": "Meta Llama 4 blog", "de": "Meta-Llama-4-Blog" }, "url": "https://ai.meta.com/blog/llama-4-multimodal-intelligence/", "checkedAt": "2026-06-09" }, { "label": { "en": "Llama 4 model card", "de": "Llama-4-Model-Karte" }, "url": "https://github.com/meta-llama/llama-models/blob/main/models/llama4/MODEL_CARD.md", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/llama-4-maverick" }, { "slug": "llama-4-scout", "name": "Llama 4 Scout", "provider": "Meta", "family": "Llama 4", "summary": { "en": "Meta's long-context Llama 4 MoE. 10M token window, 17B active / 109B total params. Open weights for codebase-scale RAG and document mining.", "de": "Metas Long-Context-Llama-4-MoE. 10M-Token-Fenster, 17B aktiv / 109B gesamt. Open Weights für Codebase-RAG und Dokumenten-Mining." }, "role": { "en": "Self-hosted route when you need entire repos or document libraries in one pass — not for lowest-latency chat.", "de": "Self-Hosting-Route, wenn ganze Repos oder Dokumentbibliotheken in einem Durchlauf nötig sind — nicht für niedrigste Chat-Latenz." }, "releaseDate": "2026-04-05", "knowledgeCutoff": "2024-08-01", "contextWindow": 10000000, "maxOutput": 16384, "modalities": [ "text", "image", "code" ], "access": [ "open-weights", "managed-cloud", "api" ], "inputPriceUSD": 0.08, "outputPriceUSD": 0.3, "pricingNote": { "en": "Typical hosted inference pricing via cloud GPU providers; weights are free under Llama 4 Community License.", "de": "Typische gehostete Inference-Preise über Cloud-GPU-Anbieter; Weights kostenlos unter Llama-4-Community-Lizenz." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Meta Llama 4 blog", "de": "Meta-Llama-4-Blog" }, "url": "https://ai.meta.com/blog/llama-4-multimodal-intelligence/", "checkedAt": "2026-06-09" }, { "label": { "en": "Llama 4 model card", "de": "Llama-4-Model-Karte" }, "url": "https://github.com/meta-llama/llama-models/blob/main/models/llama4/MODEL_CARD.md", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/llama-4-scout" }, { "slug": "longcat-2-0", "name": "LongCat-2.0", "provider": "Meituan", "family": "LongCat", "summary": { "en": "Meituan's June 2026 agentic-coding model: a 1.6T-parameter MoE with roughly 48B active parameters per token, 1M-context training, MIT licensing, and a public reveal as the model behind OpenRouter's Owl Alpha stealth route.", "de": "Meituans Agentic-Coding-Modell vom Juni 2026: ein 1,6T-Parameter-MoE mit rund 48B aktiven Parametern pro Token, 1M-Kontext-Training, MIT-Lizenz und öffentlicher Enthüllung als Modell hinter OpenRouters Stealth-Route Owl Alpha." }, "role": { "en": "Track as a near-frontier open-weights coding contender, but wait for weights and independent reruns before treating the OpenRouter/Owl Alpha adoption story as durable evidence.", "de": "Als Near-Frontier-Open-Weights-Coding-Kandidat tracken, aber auf Gewichte und unabhängige Re-Runs warten, bevor die OpenRouter/Owl-Alpha-Adoptionsstory als belastbare Evidenz zählt." }, "releaseDate": "2026-06-29", "knowledgeCutoff": "open", "contextWindow": 1000000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.75, "outputPriceUSD": 2.95, "pricingNote": { "en": "Standard LongCat API pricing reported at $0.75 input / $2.95 output per 1M tokens, with limited-time promotional pricing reported at $0.30 / $1.20 and cache-hit input free. Model weights are announced but not posted yet.", "de": "Standard-LongCat-API-Preise werden mit $0,75 Input / $2,95 Output pro 1M Tokens berichtet, zeitlich begrenzte Promo-Preise mit $0,30 / $1,20 und Cache-Hit-Input kostenlos. Modellgewichte sind angekündigt, aber noch nicht veröffentlicht." }, "pricingChecked": "2026-07-01", "status": "solid", "sources": [ { "label": { "en": "LongCat-2.0 GitHub", "de": "LongCat-2.0 GitHub" }, "url": "https://github.com/meituan-longcat/LongCat-2.0", "checkedAt": "2026-07-01" }, { "label": { "en": "LongCat-2.0 Hugging Face", "de": "LongCat-2.0 Hugging Face" }, "url": "https://huggingface.co/meituan-longcat/LongCat-2.0", "checkedAt": "2026-07-01" }, { "label": { "en": "Official LongCat-2.0 technical blog", "de": "Offizieller LongCat-2.0-Tech-Blog" }, "url": "https://longcat.chat/blog/longcat-2.0/", "checkedAt": "2026-07-01" }, { "label": { "en": "Reuters — Meituan LongCat release", "de": "Reuters — Meituan-LongCat-Release" }, "url": "https://www.reuters.com/world/china/chinas-meituan-says-new-ai-model-trained-domestic-chips-2026-06-30/", "checkedAt": "2026-07-01" }, { "label": { "en": "OpenRouter Owl Alpha model page", "de": "OpenRouter-Owl-Alpha-Modellseite" }, "url": "https://openrouter.ai/openrouter/owl-alpha", "checkedAt": "2026-07-01" } ], "html": "https://verdictpal.com/models/longcat-2-0" }, { "slug": "mimo-v2-0206", "name": "MiMo-V2-Flash (Feb 2026)", "provider": "Xiaomi", "family": "MiMo-V2-Flash (Feb", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/mimo-v2-0206" }, { "slug": "mimo-v2-5-0424", "name": "MiMo-V2.5", "provider": "Xiaomi", "family": "Xiaomi", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.14/$0.28 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.14/$0.28 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-22", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.14, "outputPriceUSD": 0.28, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/mimo-v2-5-0424" }, { "slug": "mimo-v2-5-pro-non-reasoning", "name": "MiMo-V2.5-Pro (Non-reasoning)", "provider": "Xiaomi", "family": "MiMo-V2.5-Pro (Non-reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.435/$0.87 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.435/$0.87 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-22", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.435, "outputPriceUSD": 0.87, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/mimo-v2-5-pro-non-reasoning" }, { "slug": "mimo-v2-5-pro", "name": "MiMo-V2.5-Pro", "provider": "Xiaomi", "family": "Xiaomi", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.435/$0.87 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.435/$0.87 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-22", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.435, "outputPriceUSD": 0.87, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/mimo-v2-5-pro" }, { "slug": "mimo-v2-6-flash", "name": "MiMo-V2.6-Flash", "provider": "Xiaomi", "family": "MiMo", "summary": { "en": "Xiaomi's 22 Sep 2026 open-weight MiMo-V2.6 Flash. API list $0.14 / $0.28 per 1M tokens, 1M context, 128K output. Text, image, video, and audio in; text out. No AA or Vals ledger row on this page as of 23 Sep 2026.", "de": "Xiaomis Open-Weight MiMo-V2.6 Flash vom 22. Sep. 2026. API-Liste $0,14 / $0,28 pro 1M Tokens, 1M-Kontext, 128K-Output. Text, Bild, Video und Audio rein; Text raus. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite Stand 23. Sep. 2026." }, "role": { "en": "Current Xiaomi open-weight Flash line for high-volume multimodal and tool-using calls.", "de": "Aktuelle Open-Weight-Flash-Linie von Xiaomi für High-Volume-Aufrufe mit Multimodalität und Tools." }, "releaseDate": "2026-09-22", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "open-weights" ], "inputPriceUSD": 0.14, "outputPriceUSD": 0.28, "pricingNote": { "en": "Xiaomi overseas list checked 23 Sep 2026: $0.14 input / $0.28 output per 1M tokens on a cache miss. Cache-hit input is $0.0028. Batch is half of the realtime list. Mainland China list is ¥1 / ¥2 per 1M tokens on a cache miss.", "de": "Xiaomi-Auslandsliste geprüft am 23. Sep. 2026: $0,14 Input / $0,28 Output pro 1M Tokens bei Cache-Miss. Cache-Hit-Input ist $0,0028. Batch ist die Hälfte der Echtzeitliste. Festland-China-Liste ist ¥1 / ¥2 pro 1M Tokens bei Cache-Miss." }, "pricingChecked": "2026-09-23", "status": "flagship", "sources": [ { "label": { "en": "Xiaomi MiMo-V2.6 announcement", "de": "Xiaomi-Ankündigung MiMo-V2.6" }, "url": "https://mimo.mi.com/docs/en-US/news/latest/v2-6", "checkedAt": "2026-09-23" }, { "label": { "en": "Xiaomi MiMo-V2.6-Flash model page", "de": "Xiaomi-Modellseite MiMo-V2.6-Flash" }, "url": "https://mimo.mi.com/models/en-US/mimo-v2.6-flash", "checkedAt": "2026-09-23" }, { "label": { "en": "Xiaomi MiMo batch pricing", "de": "Xiaomi-MiMo-Batch-Preise" }, "url": "https://mimo.mi.com/docs/en-US/quick-start/usage-guide/text-generation/batch-api", "checkedAt": "2026-09-23" }, { "label": { "en": "Hugging Face MiMo-V2.6 collection", "de": "Hugging-Face-Sammlung MiMo-V2.6" }, "url": "https://huggingface.co/collections/XiaomiMiMo/mimo-v26", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/mimo-v2-6-flash" }, { "slug": "mimo-v2-6-pro", "name": "MiMo-V2.6-Pro", "provider": "Xiaomi", "family": "MiMo", "summary": { "en": "Xiaomi's 22 Sep 2026 open-weight MiMo-V2.6 Pro. API list $0.435 / $0.87 per 1M tokens, 1M context, 128K output. Text, image, video, and audio in; text out. No AA or Vals ledger row on this page as of 23 Sep 2026.", "de": "Xiaomis Open-Weight MiMo-V2.6 Pro vom 22. Sep. 2026. API-Liste $0,435 / $0,87 pro 1M Tokens, 1M-Kontext, 128K-Output. Text, Bild, Video und Audio rein; Text raus. Keine AA- oder Vals-Ledger-Zeile auf dieser Seite Stand 23. Sep. 2026." }, "role": { "en": "Current Xiaomi open-weight Pro line for long-horizon coding, multimodal input, and tool use.", "de": "Aktuelle Open-Weight-Pro-Linie von Xiaomi für langlaufendes Coding, multimodalen Input und Tool-Nutzung." }, "releaseDate": "2026-09-22", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "maxOutput": 128000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app", "open-weights" ], "inputPriceUSD": 0.435, "outputPriceUSD": 0.87, "pricingNote": { "en": "Xiaomi overseas list checked 23 Sep 2026: $0.435 input / $0.87 output per 1M tokens on a cache miss. Cache-hit input is $0.0036. Batch is half of the realtime list. Mainland China list is ¥3 / ¥6 per 1M tokens on a cache miss. UltraSpeed is a separate API id with custom rate limits; this page does not copy a second USD list.", "de": "Xiaomi-Auslandsliste geprüft am 23. Sep. 2026: $0,435 Input / $0,87 Output pro 1M Tokens bei Cache-Miss. Cache-Hit-Input ist $0,0036. Batch ist die Hälfte der Echtzeitliste. Festland-China-Liste ist ¥3 / ¥6 pro 1M Tokens bei Cache-Miss. UltraSpeed ist eine eigene API-ID mit individuellen Rate-Limits; diese Seite übernimmt keine zweite USD-Liste." }, "pricingChecked": "2026-09-23", "status": "flagship", "sources": [ { "label": { "en": "Xiaomi MiMo-V2.6 announcement", "de": "Xiaomi-Ankündigung MiMo-V2.6" }, "url": "https://mimo.mi.com/docs/en-US/news/latest/v2-6", "checkedAt": "2026-09-23" }, { "label": { "en": "Xiaomi MiMo-V2.6-Pro model page", "de": "Xiaomi-Modellseite MiMo-V2.6-Pro" }, "url": "https://mimo.mi.com/models/en-US/mimo-v2.6-pro", "checkedAt": "2026-09-23" }, { "label": { "en": "Xiaomi MiMo model summary", "de": "Xiaomi-MiMo-Modellübersicht" }, "url": "https://mimo.mi.com/docs/en-US/quick-start/summary/model", "checkedAt": "2026-09-23" }, { "label": { "en": "Hugging Face MiMo-V2.6 collection", "de": "Hugging-Face-Sammlung MiMo-V2.6" }, "url": "https://huggingface.co/collections/XiaomiMiMo/mimo-v26", "checkedAt": "2026-09-23" } ], "html": "https://verdictpal.com/models/mimo-v2-6-pro" }, { "slug": "mimo-v2-flash-reasoning", "name": "MiMo-V2-Flash (Reasoning)", "provider": "Xiaomi", "family": "MiMo-V2-Flash (Reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.1/$0.3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.1/$0.3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.1, "outputPriceUSD": 0.3, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/mimo-v2-flash-reasoning" }, { "slug": "mimo-v2-omni-0327", "name": "MiMo-V2-Omni-0327", "provider": "Xiaomi", "family": "Xiaomi", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-27", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/mimo-v2-omni-0327" }, { "slug": "mimo-v2-omni", "name": "MiMo-V2-Omni", "provider": "Xiaomi", "family": "Xiaomi", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-19", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-27", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-27" } ], "html": "https://verdictpal.com/models/mimo-v2-omni" }, { "slug": "mimo-v2-pro", "name": "MiMo-V2-Pro", "provider": "Xiaomi", "family": "Xiaomi", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-03-18", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/mimo-v2-pro" }, { "slug": "minicpm-v4-6-1-3b", "name": "MiniCPM-V 4.6 1.3B", "provider": "OpenBMB", "family": "MiniCPM-V 4.6", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-11", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "image" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/minicpm-v4-6-1-3b" }, { "slug": "minicpm5-1b-non-reasoning", "name": "MiniCPM5-1B (Non-reasoning)", "provider": "OpenBMB", "family": "MiniCPM5-1B (Non-reasoning)", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-25", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/minicpm5-1b-non-reasoning" }, { "slug": "minicpm5-1b-reasoning", "name": "MiniCPM5-1B (Reasoning)", "provider": "OpenBMB", "family": "MiniCPM", "summary": { "en": "OpenBMB's 1B-class reasoning model, evaluated by Artificial Analysis on 4 Jun. It belongs in the tiny-model lane, not the frontier leaderboard lane.", "de": "OpenBMBs 1B-Reasoning-Modell, am 4. Juni von Artificial Analysis evaluiert. Es gehört in die Tiny-Model-Lane, nicht in die Frontier-Leaderboard-Lane." }, "role": { "en": "Tiny reasoning baseline for edge experiments, classroom comparisons, and cheap local routing tests.", "de": "Tiny-Reasoning-Baseline für Edge-Experimente, Unterrichtsvergleiche und günstige lokale Routing-Tests." }, "releaseDate": "2026-06-04", "knowledgeCutoff": "open", "contextWindow": 128000, "maxOutput": 8192, "modalities": [ "text", "code", "reasoning" ], "access": [ "open-weights", "on-device" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Open weights / self-hosted; no API list price captured in the local ledger yet.", "de": "Open Weights / self-hosted; noch kein API-Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/minicpm5-1b-reasoning" }, { "slug": "minicpm5-2b", "name": "MiniCPM5-2B", "provider": "OpenBMB", "family": "OpenBMB", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Small OpenBMB checkpoint tracked on Artificial Analysis. Use the ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Kleiner OpenBMB-Checkpoint auf Artificial Analysis. Nutze das Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-09-07", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-09-10", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models/minicpm5-2b", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/minicpm5-2b" }, { "slug": "minimax-m2-1", "name": "MiniMax-M2.1", "provider": "MiniMax", "family": "MiniMax", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$1.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$1.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-12-23", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/minimax-m2-1" }, { "slug": "minimax-m2-5", "name": "MiniMax M2.5", "provider": "MiniMax", "family": "MiniMax", "summary": { "en": "MiniMax's April 2026 agentic model. Superseded by M3 for open-weights work but still on leaderboards.", "de": "MiniMax' agentisches Modell April 2026. Für Open-Weights-Arbeit von M3 abgelöst, aber weiterhin auf Leaderboards." }, "role": { "en": "Prior MiniMax tier for coding agents when pinned to M2.5 endpoints.", "de": "Vorheriges MiniMax-Tier für Coding-Agenten bei gebundenen M2.5-Endpunkten." }, "releaseDate": "2026-04-01", "knowledgeCutoff": "2026-02-01", "contextWindow": 1000000, "maxOutput": 65536, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-09" }, { "label": { "en": "MiniMax", "de": "MiniMax" }, "url": "https://minimaxi.com/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/minimax-m2-5" }, { "slug": "minimax-m2-7", "name": "MiniMax M2.7", "provider": "MiniMax", "family": "MiniMax", "summary": { "en": "MiniMax's April 2026 mid-tier. Bridge between M2.5 and the M3 open-weights release.", "de": "MiniMax' Mittelklasse April 2026. Brücke zwischen M2.5 und dem M3-Open-Weights-Release." }, "role": { "en": "MiniMax mid-tier when M3 is too new for your stack or M2.5 scores aren't enough.", "de": "MiniMax-Mittelklasse, wenn M3 für den Stack zu neu ist oder M2.5-Scores nicht reichen." }, "releaseDate": "2026-04-15", "knowledgeCutoff": "2026-02-01", "contextWindow": 1000000, "maxOutput": 65536, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.35, "outputPriceUSD": 1.5, "pricingChecked": "2026-06-09", "status": "solid", "sources": [], "html": "https://verdictpal.com/models/minimax-m2-7" }, { "slug": "minimax-m3", "name": "MiniMax M3", "provider": "MiniMax", "family": "MiniMax", "summary": { "en": "MiniMax's May 2026 open-weights release. Terminal-Bench 2.0 66.0% — the #10 score globally and the strongest non-frontier-tier result on the public ledger. 1M context. Pay-as-you-go list $0.30 / $1.20 per 1M tokens on inputs up to 512K.", "de": "MiniMax' Open-Weights-Release Mai 2026. Terminal-Bench 2.0 66,0 % — global #10 und stärkstes Non-Frontier-Tier-Ergebnis im öffentlichen Ledger. 1M-Kontext. Pay-as-you-go-Liste $0,30 / $1,20 pro 1M Tokens bei Input bis 512K." }, "role": { "en": "Default open-weights model for self-hosted agent infrastructure. Cheaper than DeepSeek V4 Pro Max on output ($2 vs $1.60 — well, slightly higher), with comparable Terminal-Bench scores.", "de": "Default-Open-Weights-Modell für selbstgehostete Agent-Infrastruktur. Vergleichbar mit DeepSeek V4 Pro Max bei Terminal-Bench, mit ähnlichem Preisprofil und 1M-Kontext." }, "releaseDate": "2026-05-31", "knowledgeCutoff": "2026-03-01", "contextWindow": 1000000, "maxOutput": 65536, "modalities": [ "text", "image", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingNote": { "en": "Official pay-as-you-go list checked 28 Sep 2026. Inputs up to 512K: $0.30 input / $1.20 output per 1M tokens, after a permanent half-off from the printed rack rate. Inputs above 512K: $0.60 / $2.40. Cache reads are $0.06 on the standard tier.", "de": "Offizielle Pay-as-you-go-Liste geprüft am 28. Sep. 2026. Input bis 512K: $0,30 Input / $1,20 Output pro 1M Tokens, nach dauerhaftem Halbpreis auf dem gedruckten Listenpreis. Input über 512K: $0,60 / $2,40. Cache-Reads liegen auf der Standardstufe bei $0,06." }, "pricingChecked": "2026-09-28", "status": "solid", "sources": [ { "label": { "en": "MiniMax M3 model card", "de": "MiniMax-M3-Model-Karte" }, "url": "https://huggingface.co/MiniMaxAI/MiniMax-M3", "checkedAt": "2026-06-01" }, { "label": { "en": "MiniMax API pricing", "de": "MiniMax-API-Preise" }, "url": "https://platform.minimax.io/docs/guides/pricing-paygo", "checkedAt": "2026-09-28" }, { "label": { "en": "MiniMax M3 launch", "de": "MiniMax-M3-Launch" }, "url": "https://www.minimax.io/blog/minimax-m3", "checkedAt": "2026-09-28" }, { "label": { "en": "Terminal-Bench 2.0 leaderboard", "de": "Terminal-Bench-2.0-Leaderboard" }, "url": "https://www.tbench.ai/leaderboard/terminal-bench/2.0", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/minimax-m3" }, { "slug": "mistral-large-2-2024-07", "name": "Mistral Large 2 (2024-07)", "provider": "Mistral", "family": "Mistral Large", "summary": { "en": "Mistral's July 2024 Large 2. Predecessor to Mistral Large 3.", "de": "Mistrals Large 2 Juli 2024. Vorgänger von Mistral Large 3." }, "role": { "en": "Legacy Mistral flagship — use Mistral Large 3 for current work.", "de": "Legacy-Mistral-Flaggschiff — für aktuelle Arbeit Mistral Large 3 nutzen." }, "releaseDate": "2024-07-24", "knowledgeCutoff": "2024-06-01", "contextWindow": 128000, "maxOutput": 8192, "modalities": [ "text", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Mistral Large 2", "de": "Mistral Large 2" }, "url": "https://mistral.ai/news/mistral-large-2407/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/mistral-large-2-2024-07" }, { "slug": "mistral-large-3", "name": "Mistral Large 3", "provider": "Mistral", "family": "Mistral Large", "summary": { "en": "Mistral's December 2025 flagship MoE. 262K context, vision input, $0.50/$1.50 per 1M tokens. 675B total / 41B active parameters.", "de": "Mistrals Flaggschiff-MoE Dezember 2025. 262K-Kontext, Vision-Input, $0,50/$1,50 pro 1M Tokens. 675B gesamt / 41B aktive Parameter." }, "role": { "en": "European open-API frontier for multilingual enterprise chat, document QA, and vision+text workflows.", "de": "Europäisches Open-API-Frontier für mehrsprachigen Enterprise-Chat, Dokumenten-QA und Vision+Text-Workflows." }, "releaseDate": "2025-12-02", "knowledgeCutoff": "2025-06-01", "contextWindow": 262144, "maxOutput": 8192, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud", "open-weights" ], "inputPriceUSD": 0.5, "outputPriceUSD": 1.5, "pricingNote": { "en": "Official Mistral API list price — third-party routers may discount.", "de": "Offizieller Mistral-API-Listenpreis — Drittanbieter-Router können rabattieren." }, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Mistral model docs", "de": "Mistral-Modelldocs" }, "url": "https://docs.mistral.ai/models/", "checkedAt": "2026-06-09" }, { "label": { "en": "Mistral pricing", "de": "Mistral-Preise" }, "url": "https://mistral.ai/pricing", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/mistral-large-3" }, { "slug": "mistral-medium-3-5", "name": "Mistral Medium 3.5", "provider": "Mistral", "family": "Mistral Medium", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.5/$7.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.5/$7.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-29", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 1.5, "outputPriceUSD": 7.5, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/mistral-medium-3-5" }, { "slug": "motif-0714", "name": "Motif 3 (Beta)", "provider": "Other", "family": "Motif 3", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-07-21", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-07-21", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/motif-0714" }, { "slug": "motif-3", "name": "Motif 3", "provider": "Other", "family": "Motif", "summary": { "en": "Motif Technologies' 314B MoE (13.2B active). AA lists API pricing as $0 / $0 per 1M tokens (open weights).", "de": "314B-MoE von Motif Technologies (13,2B aktiv). AA führt API-Preise als $0 / $0 pro 1M Tokens (Open Weights)." }, "role": { "en": "Open Motif checkpoint for long-context coding and Korean-capable general work. Prefer sourced ledger rows over the technical-report table.", "de": "Offener Motif-Checkpoint für Long-Context-Coding und koreanisch-fähige Allgemeinarbeit. Bevorzuge belegte Ledger-Zeilen gegenüber der Technical-Report-Tabelle." }, "releaseDate": "2026-08-12", "knowledgeCutoff": "unpublished", "contextWindow": 262144, "modalities": [ "text", "code", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingNote": { "en": "Weights on Hugging Face (Motif-Technologies/Motif-3). AA lists $0 / $0 per 1M tokens on the 2026-08-15 snapshot. Context 262,144 tokens on the model card. Knowledge cutoff March 2026 per the technical report.", "de": "Gewichte auf Hugging Face (Motif-Technologies/Motif-3). AA führt $0 / $0 pro 1M Tokens im Snapshot vom 2026-08-15. Kontext 262.144 Tokens auf der Modellkarte. Knowledge-Cutoff März 2026 laut Technical Report." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "Hugging Face Motif-3", "de": "Hugging-Face-Motif-3" }, "url": "https://huggingface.co/Motif-Technologies/Motif-3", "checkedAt": "2026-08-15" }, { "label": { "en": "Motif 3 technical report (arXiv)", "de": "Motif-3-Technical-Report (arXiv)" }, "url": "https://arxiv.org/abs/2608.09119", "checkedAt": "2026-08-15" }, { "label": { "en": "Artificial Analysis Motif 3", "de": "Artificial-Analysis-Motif-3" }, "url": "https://artificialanalysis.ai/models/motif-3", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/motif-3" }, { "slug": "muse-glimmer", "name": "Muse Glimmer", "provider": "Meta", "family": "Muse", "summary": { "en": "Meta's 30B open-weight local-agent model (10 Aug 2026). AA hosted list price $0.325 / $1.35 per 1M tokens; weights are Apache 2.0.", "de": "Metas 30B-Open-Weight-Modell für lokale Agents (10. Aug. 2026). AA-Hosted-Listenpreis $0,325 / $1,35 pro 1M Tokens; Gewichte unter Apache 2.0." }, "role": { "en": "On-device Muse variant for always-on local agents, function calling, and local coding. Prefer sourced ledger rows over launch claims; AA scores sync from the public snapshot.", "de": "On-Device-Muse-Variante für dauerhafte lokale Agents, Function Calling und lokales Coding. Bevorzuge belegte Ledger-Zeilen gegenüber Launch-Claims; AA-Scores kommen aus dem öffentlichen Snapshot." }, "releaseDate": "2026-08-10", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0.325, "outputPriceUSD": 1.35, "pricingNote": { "en": "Weights are Apache 2.0 on Hugging Face (Muse-Glimmer-30B). AA hosted list pricing as of 2026-08-13: $0.325 input / $1.35 output per 1M tokens. Context cap is unlisted on the AA record and the Meta launch post.", "de": "Gewichte stehen unter Apache 2.0 auf Hugging Face (Muse-Glimmer-30B). AA-Hosted-Listenpreise Stand 2026-08-13: $0,325 Input / $1,35 Output pro 1M Tokens. Kontextlimit ist auf dem AA-Nachweis und im Meta-Launch-Post nicht angegeben." }, "pricingChecked": "2026-08-13", "status": "solid", "sources": [ { "label": { "en": "Meta Muse Glimmer announcement", "de": "Meta-Muse-Glimmer-Ankündigung" }, "url": "https://research.meta.ai/blog/introducing-muse-glimmer-open-agentic-model", "checkedAt": "2026-08-13" }, { "label": { "en": "Hugging Face Muse-Glimmer-30B", "de": "Hugging-Face-Muse-Glimmer-30B" }, "url": "https://huggingface.co/meta-models/Muse-Glimmer-30B", "checkedAt": "2026-08-13" }, { "label": { "en": "Artificial Analysis Muse Glimmer", "de": "Artificial-Analysis-Muse-Glimmer" }, "url": "https://artificialanalysis.ai/models/muse-glimmer", "checkedAt": "2026-08-13" } ], "html": "https://verdictpal.com/models/muse-glimmer" }, { "slug": "muse-spark-1-1", "name": "Muse Spark 1.1", "provider": "Meta", "family": "Muse Spark", "summary": { "en": "Meta's multimodal reasoning model for agentic tasks, coding, computer use, and long-running workflows, with a 1M-token managed context window.", "de": "Metas multimodales Reasoning-Modell für agentische Aufgaben, Coding, Computer-Nutzung und langlaufende Workflows, mit einem verwalteten Kontextfenster von 1 Mio. Token." }, "role": { "en": "Agentic model for parallel tool use, computer operation, multimodal work, and repository-scale coding through Meta AI or the Meta Model API.", "de": "Agentisches Modell für parallele Tool-Nutzung, Computer-Bedienung, multimodale Arbeit und Coding auf Repository-Ebene über Meta AI oder die Meta Model API." }, "releaseDate": "2026-07-09", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app" ], "inputPriceUSD": 1.25, "outputPriceUSD": 4.25, "pricingNote": { "en": "Artificial Analysis API pricing snapshot. Meta Model API remains in public preview.", "de": "API-Preissnapshot von Artificial Analysis. Die Meta Model API bleibt in der Public Preview." }, "pricingChecked": "2026-07-19", "status": "solid", "sources": [ { "label": { "en": "Meta Muse Spark 1.1 launch", "de": "Meta-Muse-Spark-1.1-Launch" }, "url": "https://ai.meta.com/blog/introducing-muse-spark-meta-model-api/", "checkedAt": "2026-07-19" }, { "label": { "en": "Muse Spark 1.1 evaluation report", "de": "Muse-Spark-1.1-Evaluationsbericht" }, "url": "https://ai.meta.com/static-resource/muse-spark-1-1-evaluation-report", "checkedAt": "2026-07-19" }, { "label": { "en": "Artificial Analysis model record", "de": "Artificial-Analysis-Modellnachweis" }, "url": "https://artificialanalysis.ai/models/muse-spark-1-1", "checkedAt": "2026-07-19" } ], "html": "https://verdictpal.com/models/muse-spark-1-1" }, { "slug": "muse-spark-1-2", "name": "Muse Spark 1.2", "provider": "Meta", "family": "Muse Spark", "summary": { "en": "Meta's August 2026 Muse Spark refresh, superseded as the current Muse Spark by 1.3 on 2 Sep 2026. AA list $1.25 / $4.25 per 1M tokens.", "de": "Metas Muse-Spark-Refresh August 2026, als aktuelles Muse Spark am 2. Sep. 2026 von 1.3 abgelöst. AA-Liste $1,25 / $4,25 pro 1M Tokens." }, "role": { "en": "Previous Muse Spark when 1.3 is not on your endpoint. Prefer sourced ledger rows over launch claims; the 56.8 figure is v4.1 and is not comparable to 1.3's live v4.3 score of 48.2.", "de": "Vorheriges Muse Spark, wenn 1.3 auf deinem Endpoint fehlt. Bevorzuge belegte Ledger-Zeilen gegenüber Launch-Claims; die 56,8 ist v4.1 und nicht mit dem live v4.3-Score 48,2 von 1.3 vergleichbar." }, "releaseDate": "2026-08-05", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app" ], "inputPriceUSD": 1.25, "outputPriceUSD": 4.25, "pricingNote": { "en": "AA list pricing for Muse Spark 1.2 (xhigh) as of 2026-08-12: $1.25 input / $4.25 output per 1M tokens. Meta Model API remains in public preview on the AA record.", "de": "AA-Listenpreise für Muse Spark 1.2 (xhigh) Stand 2026-08-12: $1.25 Input / $4.25 Output pro 1M Tokens. Die Meta Model API bleibt im AA-Nachweis in der Public Preview." }, "pricingChecked": "2026-08-12", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis Muse Spark 1.2", "de": "Artificial-Analysis-Muse-Spark-1.2" }, "url": "https://artificialanalysis.ai/models/muse-spark-1-2", "checkedAt": "2026-08-12" }, { "label": { "en": "Artificial Analysis Muse Spark 1.2 article", "de": "Artificial-Analysis-Muse-Spark-1.2-Artikel" }, "url": "https://artificialanalysis.ai/articles/muse-spark-1-2", "checkedAt": "2026-08-12" } ], "html": "https://verdictpal.com/models/muse-spark-1-2" }, { "slug": "muse-spark-1-3", "name": "Muse Spark 1.3", "provider": "Meta", "family": "Muse Spark", "summary": { "en": "Meta's 2 Sep 2026 Muse Spark refresh. Same AA list as 1.2: $1.25 / $4.25 per 1M tokens, 1M context, text / image / video input. Ships in Muse Code and the Meta Model API.", "de": "Metas Muse-Spark-Refresh vom 2. Sep. 2026. Dieselbe AA-Liste wie 1.2: $1,25 / $4,25 pro 1M Tokens, 1M-Kontext, Text-/Bild-/Video-Input. Läuft in Muse Code und der Meta Model API." }, "role": { "en": "Current Meta frontier for agentic coding and long-horizon tool use. Prefer sourced AA ledger rows over launch DeepSWE / vendor tables. Vals has no model page as of 5 Sep 2026.", "de": "Aktuelles Meta-Frontier für agentisches Coding und Long-Horizon-Tool-Use. Bevorzuge belegte AA-Ledger-Zeilen gegenüber Launch-DeepSWE- und Vendor-Tabellen. Vals hat Stand 5. Sep. 2026 keine Modellseite." }, "releaseDate": "2026-09-02", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "image", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "consumer-app" ], "inputPriceUSD": 1.25, "outputPriceUSD": 4.25, "pricingNote": { "en": "AA list pricing for Muse Spark 1.3 (max) as of 5 Sep 2026: $1.25 input / $4.25 output per 1M tokens, 88% cache-hit discount on that page. Meta's launch post does not publish a separate list. Available in Muse Code and the Meta Model API; max reasoning shipped after extra safety testing.", "de": "AA-Listenpreise für Muse Spark 1.3 (max) Stand 5. Sep. 2026: $1,25 Input / $4,25 Output pro 1M Tokens, 88 % Cache-Hit-Rabatt auf der Modellseite. Metas Launch-Post veröffentlicht keine eigene Liste. Verfügbar in Muse Code und der Meta Model API; Max-Reasoning kam nach zusätzlichen Safety-Tests." }, "pricingChecked": "2026-09-05", "status": "flagship", "sources": [ { "label": { "en": "Meta Muse Spark 1.3 announcement", "de": "Meta-Muse-Spark-1.3-Ankündigung" }, "url": "https://research.meta.ai/blog/introducing-muse-spark-1-3", "checkedAt": "2026-09-05" }, { "label": { "en": "Artificial Analysis Muse Spark 1.3", "de": "Artificial-Analysis-Muse-Spark-1.3" }, "url": "https://artificialanalysis.ai/models/muse-spark-1-3", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/muse-spark-1-3" }, { "slug": "muse-spark", "name": "Muse Spark", "provider": "Other", "family": "Muse", "summary": { "en": "Muse's tax-domain model. TaxEval v2 leader at 77.7% on Vals. Niche vertical — not a general-purpose frontier model.", "de": "Muses Steuer-Domain-Modell. TaxEval-v2-Spitze bei 77,7 % auf Vals. Nischen-Vertical — kein General-Purpose-Frontier-Modell." }, "role": { "en": "Tax and compliance workflows only — do not use as a general research or coding model.", "de": "Nur für Steuer- und Compliance-Workflows — nicht als allgemeines Research- oder Coding-Modell nutzen." }, "releaseDate": "2026-01-01", "knowledgeCutoff": "2025-06-01", "contextWindow": 128000, "maxOutput": 16384, "modalities": [ "text", "reasoning" ], "access": [ "api" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Vertical model — contact Muse or check Vals for access paths.", "de": "Vertical-Modell — Muse kontaktieren oder Vals für Zugangswege prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Vals TaxEval v2", "de": "Vals-TaxEval-v2" }, "url": "https://www.vals.ai/benchmarks/tax_eval_v2", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/muse-spark" }, { "slug": "nemotron-3-5-lightning", "name": "Nemotron 3.5 Lightning", "provider": "NVIDIA", "family": "Nemotron", "summary": { "en": "NVIDIA's 30B-class open Nemotron for local and single-GPU agent steps. AA hosted list $0.05 / $0.20 per 1M tokens.", "de": "NVIDIAs offenes Nemotron der 30B-Klasse für lokale und Single-GPU-Agent-Schritte. AA-Hosted-Liste $0,05 / $0,20 pro 1M Tokens." }, "role": { "en": "Cheap, small Nemotron for routed agent steps, not a Spark or Opus replacement. Prefer sourced ledger rows over vendor speed claims.", "de": "Günstiges, kleines Nemotron für geroutete Agent-Schritte, kein Spark- oder Opus-Ersatz. Bevorzuge belegte Ledger-Zeilen gegenüber Vendor-Speed-Claims." }, "releaseDate": "2026-08-11", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code", "tool-use" ], "access": [ "open-weights", "api", "on-device" ], "inputPriceUSD": 0.05, "outputPriceUSD": 0.2, "pricingNote": { "en": "Open weights; AA hosted list as of 2026-08-15: $0.05 input / $0.20 output per 1M tokens. Context cap is unlisted on the AA record.", "de": "Open Weights; AA-Hosted-Liste Stand 2026-08-15: $0,05 Input / $0,20 Output pro 1M Tokens. Kontextlimit ist auf dem AA-Nachweis nicht angegeben." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis Nemotron 3.5 Lightning", "de": "Artificial-Analysis-Nemotron-3.5-Lightning" }, "url": "https://artificialanalysis.ai/models/nemotron-3-5-lightning", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/nemotron-3-5-lightning" }, { "slug": "nemotron-3-nano-omni-30b-a3b", "name": "Nemotron 3 Nano Omni 30B A3B Reasoning", "provider": "NVIDIA", "family": "Nemotron 3", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.075/$0.3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.075/$0.3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-29", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.075, "outputPriceUSD": 0.3, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/nemotron-3-nano-omni-30b-a3b" }, { "slug": "nemotron-3-ultra", "name": "Nemotron 3 Ultra 550B A55B (Reasoning)", "provider": "NVIDIA", "family": "Nemotron", "summary": { "en": "NVIDIA's open reasoning MoE, evaluated by Artificial Analysis on 4 Jun. Public catalogs describe 550B total / 55B active parameters and a 262K context window; VerdictPal keeps score claims in the benchmark ledger.", "de": "NVIDIAs offenes Reasoning-MoE, am 4. Juni von Artificial Analysis evaluiert. Öffentliche Kataloge nennen 550B Gesamt- / 55B aktive Parameter und 262K Kontext; VerdictPal hält Score-Claims im Benchmark-Ledger." }, "role": { "en": "Open reasoning and orchestration model for teams already in NVIDIA's inference stack or testing large open MoEs.", "de": "Offenes Reasoning- und Orchestrierungsmodell für Teams im NVIDIA-Inference-Stack oder bei Tests grosser offener MoEs." }, "releaseDate": "2026-06-04", "knowledgeCutoff": "open", "contextWindow": 262000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "No canonical public list price captured in the local ledger yet.", "de": "Noch kein kanonischer öffentlicher Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" }, { "label": { "en": "Benchable model catalog", "de": "Benchable-Modellkatalog" }, "url": "https://benchable.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/nemotron-3-ultra" }, { "slug": "nex-n2-pro", "name": "Nex-N2-Pro", "provider": "Other", "family": "Other", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.5/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.5/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-06-02", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.5, "outputPriceUSD": 2.5, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/nex-n2-pro" }, { "slug": "north-mini-code", "name": "North Mini Code", "provider": "Cohere", "family": "North", "summary": { "en": "Cohere's small coding-focused MoE model, covered and evaluated by Artificial Analysis on 9 Jun. We track it as a code-specialist candidate rather than a general frontier model.", "de": "Coheres kleines coding-fokussiertes MoE-Modell, am 9. Juni von Artificial Analysis besprochen und evaluiert. Wir führen es als Code-Spezialisten, nicht als allgemeines Frontier-Modell." }, "role": { "en": "Focused coding model for low-cost code edit, completion, and repo-assistant experiments where a full frontier model is wasteful.", "de": "Fokussiertes Coding-Modell für günstige Code-Edit-, Completion- und Repo-Assistent-Experimente, bei denen ein volles Frontier-Modell Verschwendung wäre." }, "releaseDate": "2026-06-09", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "No canonical public list price captured in the local ledger yet.", "de": "Noch kein kanonischer öffentlicher Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/north-mini-code" }, { "slug": "nova-frontier-2", "name": "Nova Frontier 2", "provider": "Other", "family": "Nova Frontier", "summary": { "en": "Listed on Artificial Analysis with API pricing $3/$12 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $3/$12 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-06-10", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 3, "outputPriceUSD": 12, "pricingChecked": "2026-06-10", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-10" } ], "html": "https://verdictpal.com/models/nova-frontier-2" }, { "slug": "o3-pro", "name": "o3-pro", "provider": "OpenAI", "family": "o3", "summary": { "en": "Listed on Artificial Analysis with API pricing $20/$80 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $20/$80 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2025-06-10", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 20, "outputPriceUSD": 80, "pricingChecked": "2026-08-01", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/o3-pro" }, { "slug": "o3", "name": "OpenAI o3", "provider": "OpenAI", "family": "OpenAI o", "summary": { "en": "OpenAI's April 2025 reasoning model. Extended thinking for math, code, and science.", "de": "OpenAIs Reasoning-Modell April 2025. Extended Thinking für Mathe, Code und Wissenschaft." }, "role": { "en": "Reasoning-tier OpenAI pick when latency and cost are secondary to hardest single-shot problem solving.", "de": "OpenAI-Reasoning-Tier, wenn Latenz und Kosten hinter schwierigster Single-Shot-Problemlösung zurückstehen." }, "releaseDate": "2025-04-16", "knowledgeCutoff": "2024-06-01", "contextWindow": 200000, "maxOutput": 100000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Reasoning-tier pricing — check OpenAI API docs for current o-series rates.", "de": "Reasoning-Tier-Preise — aktuelle o-Serie-Tarife in den OpenAI-API-Docs prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "OpenAI o3 announcement", "de": "OpenAI-o3-Ankündigung" }, "url": "https://openai.com/index/introducing-o3-and-o4-mini/", "checkedAt": "2026-06-09" }, { "label": { "en": "LiveBench leaderboard", "de": "LiveBench-Leaderboard" }, "url": "https://livebench.ai/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/o3" }, { "slug": "openai-o1", "name": "OpenAI o1", "provider": "OpenAI", "family": "OpenAI o", "summary": { "en": "OpenAI's September 2024 reasoning model. Predecessor to o3 and GPT-5 reasoning tiers.", "de": "OpenAIs Reasoning-Modell September 2024. Vorgänger von o3 und GPT-5-Reasoning-Tiers." }, "role": { "en": "Legacy reasoning tier — prefer o3 or GPT-5.5 for new math and code work.", "de": "Legacy-Reasoning-Tier — für neue Mathe- und Code-Arbeit o3 oder GPT-5.5 bevorzugen." }, "releaseDate": "2024-09-12", "knowledgeCutoff": "2023-10-01", "contextWindow": 200000, "maxOutput": 100000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "Legacy o1 pricing — check OpenAI docs for archival access.", "de": "Legacy-o1-Preise — OpenAI-Docs für Archiv-Zugang prüfen." }, "pricingChecked": "2026-06-09", "status": "lightweight", "sources": [ { "label": { "en": "Learning to reason with LLMs", "de": "Learning to reason with LLMs" }, "url": "https://openai.com/index/learning-to-reason-with-llms/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/openai-o1" }, { "slug": "qwen-3-6-plus", "name": "Qwen 3.6 Plus", "provider": "Alibaba", "family": "Qwen 3", "summary": { "en": "Alibaba's April 2026 Qwen tier. Step below Qwen 3.7 Max on the public ledger.", "de": "Alibabas Qwen-Tier April 2026. Eine Stufe unter Qwen 3.7 Max im öffentlichen Ledger." }, "role": { "en": "Alibaba coding route when 3.7 Max pricing is too high for your volume.", "de": "Alibaba-Coding-Route, wenn 3.7-Max-Preise für das Volumen zu hoch sind." }, "releaseDate": "2026-04-01", "knowledgeCutoff": "2026-02-01", "contextWindow": 256000, "maxOutput": 32768, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.5, "outputPriceUSD": 2, "pricingChecked": "2026-06-09", "status": "solid", "sources": [ { "label": { "en": "Listed on the model atlas. Sourced scores that still have a public explainer are on the ledger.", "de": "Im Modell-Atlas gelistet. Belegte Scores mit öffentlicher Erklärseite stehen im Ledger." }, "url": "https://www.swebench.com/", "checkedAt": "2026-06-09" }, { "label": { "en": "Qwen", "de": "Qwen" }, "url": "https://qwenlm.github.io/", "checkedAt": "2026-06-09" } ], "html": "https://verdictpal.com/models/qwen-3-6-plus" }, { "slug": "qwen-3-7-max", "name": "Qwen 3.7 Max", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's May 2026 flagship open-weights model, matching frontier reasoning at open-weight price points.", "de": "Alibabas Open-Weights-Flaggschiff Mai 2026, das Frontier-Reasoning zu Open-Weight-Preisen bietet." }, "role": { "en": "Default model for cost-sensitive deployments, multilingual applications (especially Chinese), and teams that need open weights with frontier-tier scores.", "de": "Default-Modell für kostenkritische Deployments, mehrsprachige Anwendungen (insbesondere Chinesisch) und Teams, die Open Weights mit Frontier-Scores brauchen." }, "releaseDate": "2026-05-20", "knowledgeCutoff": "2026-02-01", "contextWindow": 128000, "maxOutput": 32768, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights", "managed-cloud" ], "inputPriceUSD": 0.4, "outputPriceUSD": 1.2, "pricingNote": { "en": "Open weights are available under the Tongyi Qianwen license. API pricing is one of the lowest among frontier models.", "de": "Open Weights unter der Tongyi-Qianwen-Lizenz verfügbar. API-Preise gehören zu den niedrigsten unter Frontier-Modellen." }, "pricingChecked": "2026-06-01", "status": "flagship", "sources": [ { "label": { "en": "Qwen 3.7 Max model card", "de": "Qwen-3.7-Max-Model-Karte" }, "url": "https://huggingface.co/Qwen/Qwen3.7-Max", "checkedAt": "2026-06-01" }, { "label": { "en": "Alibaba Cloud pricing", "de": "Alibaba-Cloud-Preise" }, "url": "https://www.alibabacloud.com/en/product/machine-learning", "checkedAt": "2026-06-01" }, { "label": { "en": "Artificial Analysis", "de": "Artificial Analysis" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-01" } ], "html": "https://verdictpal.com/models/qwen-3-7-max" }, { "slug": "qwen-3-7-plus", "name": "Qwen 3.7 Plus", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's cost-effective Qwen 3.7 tier, evaluated by Artificial Analysis on 2 Jun. Public catalogs describe multimodal input and a 1M context window; Qwen 3.7 Max remains the flagship tool.", "de": "Alibabas kosteneffizientes Qwen-3.7-Tier, am 2. Juni von Artificial Analysis evaluiert. Öffentliche Kataloge nennen multimodalen Input und ein 1M-Kontextfenster; Qwen 3.7 Max bleibt das Flaggschiff-Tool." }, "role": { "en": "Cheaper Qwen route for multilingual and multimodal workloads that do not need the Max tier.", "de": "Günstigere Qwen-Route für mehrsprachige und multimodale Workloads, die das Max-Tier nicht brauchen." }, "releaseDate": "2026-06-02", "knowledgeCutoff": "open", "contextWindow": 1000000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "No canonical public list price captured in the local ledger yet.", "de": "Noch kein kanonischer öffentlicher Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" }, { "label": { "en": "Benchable model catalog", "de": "Benchable-Modellkatalog" }, "url": "https://benchable.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/qwen-3-7-plus" }, { "slug": "qwen3-5-122b-a10b", "name": "Qwen3.5 122B A10B (Reasoning)", "provider": "Alibaba", "family": "Qwen3.5 122B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.4/$3.2 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.4/$3.2 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-24", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.4, "outputPriceUSD": 3.2, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/qwen3-5-122b-a10b" }, { "slug": "qwen3-5-27b", "name": "Qwen3.5 27B (Reasoning)", "provider": "Alibaba", "family": "Qwen3.5 27B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$2.4 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$2.4 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-24", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 2.4, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/qwen3-5-27b" }, { "slug": "qwen3-5-397b-a17b-non-reasoning", "name": "Qwen3.5 397B A17B (Non-reasoning)", "provider": "Alibaba", "family": "Qwen3.5 397B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.6/$3.6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.6/$3.6 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.6, "outputPriceUSD": 3.6, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/qwen3-5-397b-a17b-non-reasoning" }, { "slug": "qwen3-5-397b-a17b", "name": "Qwen3.5 397B A17B (Reasoning)", "provider": "Alibaba", "family": "Qwen3.5 397B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.6/$3.6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.6/$3.6 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-02-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.6, "outputPriceUSD": 3.6, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/qwen3-5-397b-a17b" }, { "slug": "qwen3-6-27b-non-reasoning", "name": "Qwen3.6 27B (Non-reasoning)", "provider": "Alibaba", "family": "Qwen3.6 27B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.6/$3.6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.6/$3.6 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-22", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.6, "outputPriceUSD": 3.6, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/qwen3-6-27b-non-reasoning" }, { "slug": "qwen3-6-27b", "name": "Qwen3.6 27B (Reasoning)", "provider": "Alibaba", "family": "Qwen3.6 27B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.6/$3.6 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.6/$3.6 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-22", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.6, "outputPriceUSD": 3.6, "pricingChecked": "2026-07-26", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-26" } ], "html": "https://verdictpal.com/models/qwen3-6-27b" }, { "slug": "qwen3-6-35b-a3b-non-reasoning", "name": "Qwen3.6 35B A3B (Non-reasoning)", "provider": "Alibaba", "family": "Qwen3.6 35B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.375/$2.25 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.375/$2.25 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.375, "outputPriceUSD": 2.25, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/qwen3-6-35b-a3b-non-reasoning" }, { "slug": "qwen3-6-35b-a3b", "name": "Qwen3.6 35B A3B (Reasoning)", "provider": "Alibaba", "family": "Qwen3.6 35B", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.248/$1.485 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.248/$1.485 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-16", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.248, "outputPriceUSD": 1.485, "pricingChecked": "2026-08-01", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-01" } ], "html": "https://verdictpal.com/models/qwen3-6-35b-a3b" }, { "slug": "qwen3-6-max", "name": "Qwen3.6 Max Preview", "provider": "Alibaba", "family": "Qwen3.6 Max", "summary": { "en": "Listed on Artificial Analysis with API pricing $1.3/$7.8 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $1.3/$7.8 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-20", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 1.3, "outputPriceUSD": 7.8, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/qwen3-6-max" }, { "slug": "qwen3-6-plus", "name": "Qwen3.6 Plus", "provider": "Alibaba", "family": "Qwen3.6 Plus", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.5/$3 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.5/$3 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-04-02", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 0.5, "outputPriceUSD": 3, "pricingChecked": "2026-07-21", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-21" } ], "html": "https://verdictpal.com/models/qwen3-6-plus" }, { "slug": "qwen3-7-max", "name": "Qwen3.7 Max", "provider": "Alibaba", "family": "Qwen3.7 Max", "summary": { "en": "Listed on Artificial Analysis with API pricing $2.5/$7.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $2.5/$7.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-19", "knowledgeCutoff": "open", "contextWindow": 128000, "modalities": [ "text", "code", "reasoning" ], "access": [ "api" ], "inputPriceUSD": 2.5, "outputPriceUSD": 7.5, "pricingChecked": "2026-07-09", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-09" } ], "html": "https://verdictpal.com/models/qwen3-7-max" }, { "slug": "qwen3-8-2-4t-a95b", "name": "Qwen3.8 2.4T A95B", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's open-weight Qwen-Max-class MoE (12 Aug 2026): 2.4T total / 95B active. AA list price $2 / $6 per 1M tokens.", "de": "Alibabas Open-Weight-Qwen-Max-Klasse-MoE (12. Aug. 2026): 2,4T gesamt / 95B aktiv. AA-Listenpreis $2 / $6 pro 1M Tokens." }, "role": { "en": "Self-hostable Qwen3.8 checkpoint for long-context and agentic work. Hosted Qwen3.8 Max adds vision and a non-thinking mode. Prefer sourced ledger rows over vendor eval tables.", "de": "Selbst hostbarer Qwen3.8-Checkpoint für langen Kontext und agentische Arbeit. Gehostetes Qwen3.8 Max ergänzt Vision und einen Non-Thinking-Modus. Bevorzuge belegte Ledger-Zeilen gegenüber Vendor-Eval-Tabellen." }, "releaseDate": "2026-08-12", "knowledgeCutoff": "unpublished", "contextWindow": 262144, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingNote": { "en": "Open weights on Hugging Face (Qwen3.8-2.4T-A95B). AA hosted list as of 2026-08-15: $2 input / $6 output per 1M tokens. Native context 262,144 tokens, extensible toward 1M on the vendor card. Text-only open checkpoint; thinking mode is required in the published interface.", "de": "Open Weights auf Hugging Face (Qwen3.8-2.4T-A95B). AA-Hosted-Liste Stand 2026-08-15: $2 Input / $6 Output pro 1M Tokens. Nativer Kontext 262.144 Tokens, auf der Vendor-Karte Richtung 1M erweiterbar. Open-Checkpoint nur Text; Thinking-Modus ist in der veröffentlichten Schnittstelle Pflicht." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "Hugging Face Qwen3.8-2.4T-A95B", "de": "Hugging-Face-Qwen3.8-2.4T-A95B" }, "url": "https://huggingface.co/Qwen/Qwen3.8-2.4T-A95B", "checkedAt": "2026-08-15" }, { "label": { "en": "Artificial Analysis Qwen3.8 2.4T A95B", "de": "Artificial-Analysis-Qwen3.8-2.4T-A95B" }, "url": "https://artificialanalysis.ai/models/qwen3-8-2-4t-a95b", "checkedAt": "2026-08-15" }, { "label": { "en": "Alibaba Cloud Qwen3.8-Max blog", "de": "Alibaba-Cloud-Qwen3.8-Max-Blog" }, "url": "https://www.alibabacloud.com/blog/qwen3-8-max-a-new-bar-for-coding-and-cowork_603421", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/qwen3-8-2-4t-a95b" }, { "slug": "qwen3-8-27b", "name": "Qwen3.8 27B", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's dense open-weight Qwen3.8 sibling (14 Aug 2026). AA hosted list $0.50 / $3 per 1M tokens. Apache 2.0 weights on Hugging Face.", "de": "Alibabas dichtes Open-Weight-Qwen3.8-Geschwister (14. Aug. 2026). AA-Hosted-Liste $0,50 / $3 pro 1M Tokens. Apache-2.0-Weights auf Hugging Face." }, "role": { "en": "Self-hostable Qwen3.8 dense checkpoint for local and single-GPU runs. Hosted Qwen3.8 Max is a different product. Prefer sourced ledger rows over vendor SWE/OSWorld tables; AA scores sync from the public snapshot.", "de": "Selbst hostbarer dichter Qwen3.8-Checkpoint für lokale und Single-GPU-Läufe. Gehostetes Qwen3.8 Max ist ein anderes Produkt. Bevorzuge belegte Ledger-Zeilen gegenüber Vendor-SWE-/OSWorld-Tabellen; AA-Scores kommen aus dem öffentlichen Snapshot." }, "releaseDate": "2026-08-14", "knowledgeCutoff": "unpublished", "contextWindow": 262144, "modalities": [ "text", "image", "video", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0.5, "outputPriceUSD": 3, "pricingNote": { "en": "Open weights on Hugging Face (Qwen3.8-27B, Apache 2.0). AA hosted list as of 2026-08-24: $0.50 input / $3 output per 1M tokens. Native context 262,144 tokens on the vendor card, extensible toward 1M. Effort variants (xhigh / medium / low) collapse onto this page.", "de": "Open Weights auf Hugging Face (Qwen3.8-27B, Apache 2.0). AA-Hosted-Liste Stand 2026-08-24: $0,50 Input / $3 Output pro 1M Tokens. Nativer Kontext 262.144 Tokens auf der Vendor-Karte, Richtung 1M erweiterbar. Effort-Varianten (xhigh / medium / low) laufen auf dieser Seite zusammen." }, "pricingChecked": "2026-08-24", "status": "solid", "sources": [ { "label": { "en": "Hugging Face Qwen3.8-27B", "de": "Hugging-Face-Qwen3.8-27B" }, "url": "https://huggingface.co/Qwen/Qwen3.8-27B", "checkedAt": "2026-08-24" }, { "label": { "en": "Alibaba Cloud Qwen3.8-27B blog", "de": "Alibaba-Cloud-Qwen3.8-27B-Blog" }, "url": "https://www.alibabacloud.com/blog/alibaba-unveils-qwen3-8-27b-and-releases-weights-of-qwen3-8-flagship-model_603463", "checkedAt": "2026-08-24" }, { "label": { "en": "Artificial Analysis Qwen3.8 27B", "de": "Artificial-Analysis-Qwen3.8-27B" }, "url": "https://artificialanalysis.ai/models/qwen3-8-27b", "checkedAt": "2026-08-24" } ], "html": "https://verdictpal.com/models/qwen3-8-27b" }, { "slug": "qwen3-8-flash-next", "name": "Qwen3.8-Flash-Next", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's 26 Aug 2026 Flash-Next sibling in the Qwen3.8 line. AA hosted list $0.15 / $0.47 per 1M tokens — cheaper and faster than Qwen3.8 27B, below hosted Qwen3.8 Max.", "de": "Alibabas Flash-Next-Geschwister in der Qwen3.8-Linie vom 26. Aug. 2026. AA-Hosted-Liste $0,15 / $0,47 pro 1M Tokens — günstiger und schneller als Qwen3.8 27B, unter gehostetem Qwen3.8 Max." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-08-26", "knowledgeCutoff": "unpublished", "contextWindow": 262144, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "open-weights" ], "inputPriceUSD": 0.15, "outputPriceUSD": 0.47, "pricingChecked": "2026-09-01", "status": "solid", "sources": [ { "label": { "en": "Qwen3.8-Flash-Next announcement", "de": "Qwen3.8-Flash-Next-Ankündigung" }, "url": "https://qwen.ai/blog?id=qwen3.8-flash-next", "checkedAt": "2026-09-01" }, { "label": { "en": "Hugging Face Qwen3.8-Flash-Next", "de": "Hugging-Face-Qwen3.8-Flash-Next" }, "url": "https://huggingface.co/Qwen/Qwen3.8-Flash-Next", "checkedAt": "2026-09-01" }, { "label": { "en": "Artificial Analysis Qwen3.8-Flash-Next", "de": "Artificial-Analysis-Qwen3.8-Flash-Next" }, "url": "https://artificialanalysis.ai/models/qwen3-8-flash-next", "checkedAt": "2026-09-01" } ], "html": "https://verdictpal.com/models/qwen3-8-flash-next" }, { "slug": "qwen3-8-livetranslate-flash-realtime", "name": "Qwen3.8-LiveTranslate", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's 19 Sep 2026 real-time interpretation model. Audio and image in; text and audio out. 53,248-token context (49,152 in / 4,096 out). International Model Studio audio list $7.50 / $30.00 per 1M tokens. Understands 60 languages; speaks 29.", "de": "Alibabas Echtzeit-Dolmetschmodell vom 19. Sep. 2026. Audio und Bild rein; Text und Audio raus. 53.248-Token-Kontext (49.152 rein / 4.096 raus). Internationale Model-Studio-Audio-Liste $7,50 / $30,00 pro 1M Tokens. Versteht 60 Sprachen; spricht 29." }, "role": { "en": "Current Qwen simultaneous-interpretation workhorse for live meetings and bilingual subtitles — not the text-out Omni-Flash ingest line.", "de": "Aktuelles Qwen-Simultandolmetsch-Arbeitspferd für Live-Meetings und zweisprachige Untertitel — nicht die textlastige Omni-Flash-Ingest-Linie." }, "releaseDate": "2026-09-19", "knowledgeCutoff": "unpublished", "contextWindow": 53248, "maxOutput": 4096, "modalities": [ "text", "image", "audio" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 7.5, "outputPriceUSD": 30, "pricingNote": { "en": "Alibaba Cloud Model Studio international list (Singapore, docs last updated 19 Sep 2026): audio input $7.50 / image input $0.55 / text output $20.00 / audio output $30.00 per 1M tokens. Chinese mainland (Beijing): $5.653 / $0.466 / $14.133 / $22.613. RPM 10 and TPM 100,000 in both regions. Call qwen3.8-livetranslate-flash-realtime on the WebSocket Realtime API. Function calling, structured outputs, and context caching are unsupported.", "de": "Alibaba-Cloud-Model-Studio internationale Liste (Singapur, Docs zuletzt 19. Sep. 2026): Audio-Input $7,50 / Bild-Input $0,55 / Text-Output $20,00 / Audio-Output $30,00 pro 1M Tokens. Chinesisches Festland (Peking): $5,653 / $0,466 / $14,133 / $22,613. RPM 10 und TPM 100.000 in beiden Regionen. Als qwen3.8-livetranslate-flash-realtime über die WebSocket-Realtime-API aufrufen. Function Calling, Structured Outputs und Context Caching sind nicht unterstützt." }, "pricingChecked": "2026-09-20", "status": "solid", "sources": [ { "label": { "en": "Alibaba Model Studio Qwen3.8-LiveTranslate-Flash-Realtime", "de": "Alibaba Model Studio Qwen3.8-LiveTranslate-Flash-Realtime" }, "url": "https://www.alibabacloud.com/help/en/model-studio/qwen3-8-livetranslate-flash-realtime", "checkedAt": "2026-09-20" }, { "label": { "en": "QwenCloud Qwen3.8-LiveTranslate-Flash-Realtime", "de": "QwenCloud Qwen3.8-LiveTranslate-Flash-Realtime" }, "url": "https://www.qwencloud.com/models/qwen3.8-livetranslate-flash-realtime", "checkedAt": "2026-09-20" }, { "label": { "en": "QwenCloud model releases", "de": "QwenCloud-Modellreleases" }, "url": "https://docs.qwencloud.com/changelog/models", "checkedAt": "2026-09-20" }, { "label": { "en": "Qwen3.8-LiveTranslate announcement", "de": "Qwen3.8-LiveTranslate-Ankündigung" }, "url": "https://www.alibabacloud.com/blog/qwen3-8-livetranslate-names-the-speaker--carries-the-meaning-_603581", "checkedAt": "2026-09-20" } ], "html": "https://verdictpal.com/models/qwen3-8-livetranslate-flash-realtime" }, { "slug": "qwen3-8-max", "name": "Qwen3.8 Max", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's August 2026 Max-tier flagship. API list $2 / $6 per 1M tokens on that snapshot.", "de": "Alibabas Max-Flaggschiff August 2026. API-Liste $2 / $6 pro 1M Tokens in diesem Snapshot." }, "role": { "en": "Current Alibaba Cloud Max model for long-context and agentic API work. Prefer sourced ledger rows over vendor eval tables; AA scores sync from the public snapshot.", "de": "Aktuelles Alibaba-Cloud-Max-Modell für langen Kontext und agentische API-Arbeit. Bevorzuge belegte Ledger-Zeilen gegenüber Vendor-Eval-Tabellen; AA-Scores kommen aus dem öffentlichen Snapshot." }, "releaseDate": "2026-08-03", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "image", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 2, "outputPriceUSD": 6, "pricingNote": { "en": "International Model Studio list re-checked 28 Sep 2026: $2 input / $6 output per 1M tokens, including the dated 2 Sep snapshot. The Beijing standard row is lower. A separate Beijing fast mode is higher and is not this list. The Max tier remains a hosted API.", "de": "Internationale Model-Studio-Liste erneut geprüft am 28. Sep. 2026: $2 Input / $6 Output pro 1M Tokens, einschliesslich des datierten Snapshots vom 2. Sep. Die Standardzeile Peking liegt niedriger. Ein separater Fast-Mode in Peking liegt höher und ist nicht diese Liste. Die Max-Stufe bleibt eine gehostete API." }, "pricingChecked": "2026-09-28", "status": "flagship", "sources": [ { "label": { "en": "Alibaba Cloud Qwen3.8-Max blog", "de": "Alibaba-Cloud-Qwen3.8-Max-Blog" }, "url": "https://www.alibabacloud.com/blog/qwen3-8-max-a-new-bar-for-coding-and-cowork_603421", "checkedAt": "2026-08-12" }, { "label": { "en": "Alibaba Cloud Model Studio", "de": "Alibaba-Cloud-Model-Studio" }, "url": "https://modelstudio.alibabacloud.com/", "checkedAt": "2026-08-12" }, { "label": { "en": "Alibaba Model Studio pricing", "de": "Alibaba-Model-Studio-Preise" }, "url": "https://www.alibabacloud.com/help/en/model-studio/model-pricing", "checkedAt": "2026-09-28" }, { "label": { "en": "Artificial Analysis Qwen3.8 Max", "de": "Artificial-Analysis-Qwen3.8-Max" }, "url": "https://artificialanalysis.ai/models/qwen3-8-max", "checkedAt": "2026-09-10" } ], "html": "https://verdictpal.com/models/qwen3-8-max" }, { "slug": "qwen3-8-omni-flash", "name": "Qwen3.8-Omni-Flash", "provider": "Alibaba", "family": "Qwen", "summary": { "en": "Alibaba's 17 Sep 2026 omnimodal Flash API model. Text, image, audio, and video in; text out. 1M context, 131K max output. International Model Studio list $0.15 / $0.47 per 1M tokens (cache-hit input $0.016).", "de": "Alibabas omnimodales Flash-API-Modell vom 17. Sep. 2026. Text, Bild, Audio und Video rein; Text raus. 1M-Kontext, 131K max. Internationale Model-Studio-Liste $0,15 / $0,47 pro 1M Tokens (Cache-Hit-Input $0,016)." }, "role": { "en": "Current Qwen omni Flash for meeting audio, video, and mixed-media ingest with tool use — not the text-only Flash-Next or Max coding line.", "de": "Aktuelles Qwen-Omni-Flash für Meeting-Audio, Video und Mixed-Media-Ingest mit Tool-Use — nicht die textlastige Flash-Next- oder Max-Coding-Linie." }, "releaseDate": "2026-09-17", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "maxOutput": 131072, "modalities": [ "text", "image", "audio", "video", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.15, "outputPriceUSD": 0.47, "pricingNote": { "en": "Alibaba Cloud Model Studio international list (Singapore and other non-mainland regions): $0.15 input / $0.016 cache-hit input / $0.47 output per 1M tokens. Chinese mainland: $0.113 / $0.014 / $0.382. One published input rate across modalities; Alibaba still applies modality token-conversion rules. Call Qwen3.8-Omni-Flash on Model Studio. Thinking is on by default.", "de": "Alibaba-Cloud-Model-Studio internationale Liste (Singapur und andere Non-Mainland-Regionen): $0,15 Input / $0,016 Cache-Hit-Input / $0,47 Output pro 1M Tokens. Chinesisches Festland: $0,113 / $0,014 / $0,382. Ein veröffentlichter Input-Tarif über Modalitäten; Alibaba wendet trotzdem Modalitäts-Token-Umrechnung an. Auf Model Studio als Qwen3.8-Omni-Flash aufrufen. Thinking ist standardmässig an." }, "pricingChecked": "2026-09-18", "status": "solid", "sources": [ { "label": { "en": "Alibaba Model Studio Qwen3.8-Omni-Flash", "de": "Alibaba Model Studio Qwen3.8-Omni-Flash" }, "url": "https://www.alibabacloud.com/help/en/model-studio/qwen3-8-omni-flash", "checkedAt": "2026-09-18" }, { "label": { "en": "Alibaba Model Studio model pricing", "de": "Alibaba-Model-Studio-Modellpreise" }, "url": "https://www.alibabacloud.com/help/en/model-studio/model-pricing", "checkedAt": "2026-09-18" } ], "html": "https://verdictpal.com/models/qwen3-8-omni-flash" }, { "slug": "qwen3-max-thinking", "name": "Qwen3 Max Thinking", "provider": "Alibaba", "family": "Qwen3 Max", "summary": { "en": "Listed on Artificial Analysis with API pricing $0/$0 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0/$0 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-01-26", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingChecked": "2026-08-09", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-08-09" } ], "html": "https://verdictpal.com/models/qwen3-max-thinking" }, { "slug": "ring-2-6-1t", "name": "Ring-2.6-1T", "provider": "Other", "family": "Other", "summary": { "en": "Listed on Artificial Analysis with API pricing $0.3/$2.5 per 1M tokens. Benchmark rows sync from the AA snapshot — see the ledger on this page.", "de": "Auf Artificial Analysis gelistet mit API-Preisen $0.3/$2.5 pro 1M Tokens. Benchmark-Zeilen werden aus dem AA-Snapshot synchronisiert — siehe Ledger auf dieser Seite." }, "role": { "en": "Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.", "de": "Frontier-Modell auf Artificial Analysis. Nutze das Benchmark-Ledger unten für belegte Scores; wir leiten keine Zahlen über veröffentlichte Zeilen hinaus ab." }, "releaseDate": "2026-05-08", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text" ], "access": [ "api" ], "inputPriceUSD": 0.3, "outputPriceUSD": 2.5, "pricingChecked": "2026-07-31", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-07-31" } ], "html": "https://verdictpal.com/models/ring-2-6-1t" }, { "slug": "solar-open2-250b", "name": "Solar Open2 250B", "provider": "Other", "family": "Solar", "summary": { "en": "Upstage Solar Open 2, a 250B-A15B MoE for long-horizon agent work. AA lists $0 / $0 per 1M tokens.", "de": "Upstage Solar Open 2, ein 250B-A15B-MoE für Long-Horizon-Agent-Arbeit. AA führt $0 / $0 pro 1M Tokens." }, "role": { "en": "Open Solar checkpoint aimed at million-token agent trajectories. Prefer sourced ledger rows over the technical-report table.", "de": "Offener Solar-Checkpoint für Million-Token-Agent-Trajectories. Bevorzuge belegte Ledger-Zeilen gegenüber der Technical-Report-Tabelle." }, "releaseDate": "2026-08-12", "knowledgeCutoff": "unpublished", "contextWindow": 1000000, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "open-weights", "api" ], "inputPriceUSD": 0, "outputPriceUSD": 0, "pricingNote": { "en": "AA lists $0 / $0 per 1M tokens on the 2026-08-15 snapshot (open weights). Technical report describes a 1M-token hybrid-attention window. Hosted Solar Pro 4 is a separate paid Upstage product.", "de": "AA führt $0 / $0 pro 1M Tokens im Snapshot vom 2026-08-15 (Open Weights). Der Technical Report beschreibt ein 1M-Token-Hybrid-Attention-Fenster. Gehostetes Solar Pro 4 ist ein separates bezahltes Upstage-Produkt." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "Solar Open 2 technical report (arXiv)", "de": "Solar-Open-2-Technical-Report (arXiv)" }, "url": "https://arxiv.org/abs/2607.20062", "checkedAt": "2026-08-15" }, { "label": { "en": "Artificial Analysis Solar Open2 250B", "de": "Artificial-Analysis-Solar-Open2-250B" }, "url": "https://artificialanalysis.ai/models/solar-open2-250b", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/solar-open2-250b" }, { "slug": "solar-pro4", "name": "Solar Pro 4", "provider": "Other", "family": "Solar", "summary": { "en": "Upstage hosted Solar Pro 4. AA list $0.30 / $1.20 per 1M tokens.", "de": "Gehostetes Upstage Solar Pro 4. AA-Liste $0,30 / $1,20 pro 1M Tokens." }, "role": { "en": "Paid Solar API workhorse. Open Solar Open2 250B is the weight-release sibling. Prefer sourced ledger rows.", "de": "Bezahltes Solar-API-Arbeitspferd. Open Solar Open2 250B ist das Weight-Release-Geschwister. Bevorzuge belegte Ledger-Zeilen." }, "releaseDate": "2026-08-06", "knowledgeCutoff": "unpublished", "contextWindow": null, "modalities": [ "text", "code", "tool-use", "reasoning" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": 0.3, "outputPriceUSD": 1.2, "pricingNote": { "en": "AA hosted list as of 2026-08-15: $0.30 input / $1.20 output per 1M tokens. Context cap is unlisted on the AA record.", "de": "AA-Hosted-Liste Stand 2026-08-15: $0,30 Input / $1,20 Output pro 1M Tokens. Kontextlimit ist auf dem AA-Nachweis nicht angegeben." }, "pricingChecked": "2026-08-15", "status": "solid", "sources": [ { "label": { "en": "Artificial Analysis Solar Pro 4", "de": "Artificial-Analysis-Solar-Pro-4" }, "url": "https://artificialanalysis.ai/models/solar-pro4", "checkedAt": "2026-08-15" } ], "html": "https://verdictpal.com/models/solar-pro4" }, { "slug": "step-3-7-flash", "name": "Step 3.7 Flash", "provider": "StepFun", "family": "Step", "summary": { "en": "StepFun's high-efficiency multimodal Flash model, evaluated by Artificial Analysis on 1 Jun. Public catalogs describe image/video input and a 256K context window.", "de": "StepFuns hocheffizientes multimodales Flash-Modell, am 1. Juni von Artificial Analysis evaluiert. Öffentliche Kataloge nennen Bild-/Video-Input und ein 256K-Kontextfenster." }, "role": { "en": "Efficiency-first multimodal model for routing, visual extraction, and cheap agent steps before escalating to a larger model.", "de": "Effizienzorientiertes multimodales Modell für Routing, visuelle Extraktion und günstige Agent-Schritte vor Eskalation zu einem grösseren Modell." }, "releaseDate": "2026-06-01", "knowledgeCutoff": "open", "contextWindow": 256000, "modalities": [ "text", "image", "video", "code", "tool-use" ], "access": [ "api", "managed-cloud" ], "inputPriceUSD": null, "outputPriceUSD": null, "pricingNote": { "en": "No canonical public list price captured in the local ledger yet.", "de": "Noch kein kanonischer öffentlicher Listenpreis im lokalen Ledger erfasst." }, "pricingChecked": "2026-06-14", "status": "lightweight", "sources": [ { "label": { "en": "Artificial Analysis models index", "de": "Artificial-Analysis-Modellindex" }, "url": "https://artificialanalysis.ai/models", "checkedAt": "2026-06-14" }, { "label": { "en": "Benchable model catalog", "de": "Benchable-Modellkatalog" }, "url": "https://benchable.ai/models", "checkedAt": "2026-06-14" } ], "html": "https://verdictpal.com/models/step-3-7-flash" } ] }