VerdictPal
Join

Models / Anthropic

Claude Opus 5 (Adaptive Reasoning, Medium Effort)

Claude Opus · Released 2026-07-24

AA Intelligence Index 56.3. 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.

Frontier model tracked on Artificial Analysis. Use the benchmark ledger below for sourced scores; we do not infer numbers beyond published rows.

#5 of 184 on AA Index · current snapshot
45 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite63
Context128K
Input / 1M$5.0
Output / 1M$25
Knowledge cutoffopen
Output speed52 t/s
TTFT (default API)6.40s
Statussolid

Family profile

Best published score in each covered benchmark family.

78/100 avg
Reasoning
92
Coding
52
Agentic
86
Long context
82

4 tested benchmark families

Benchmark placements

Where Claude Opus 5 (Adaptive Reasoning, Medium Effort) places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
Humanity's Last ExamReasoning#7/ 18551Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#9/ 18845Artificial Analysis2026-09-10
Terminal-BenchAgentic#10/ 18986Artificial Analysis2026-09-10
AA-LCRLong context#16/ 18282Artificial Analysis2026-09-10
AA time to first tokenPerformance#24/ 190Artificial Analysis2026-09-10
τ³-Bench BankingAgentic#25/ 11139Artificial Analysis2026-09-10
GPQA DiamondReasoning#28/ 19092Artificial Analysis2026-09-10
SciCodeCoding#33/ 9452Artificial Analysis2026-09-10
AA output speedPerformance#74/ 190Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Claude Opus 5 (Adaptive Reasoning, Medium Effort) shows up, it's highlighted.

ReasoningClaude Opus 5 (Adaptive Reasoning, Medium Effort) · 92
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
CodingClaude Opus 5 (Adaptive Reasoning, Medium Effort) · 52
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticClaude Opus 5 (Adaptive Reasoning, Medium Effort) · 86
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextClaude Opus 5 (Adaptive Reasoning, Medium Effort) · 82
  1. 1
    Kimi K3
    89
  2. 2
    Claude Fable 5.1
    85
  3. 3
    GPT-5.5 (high)
    84
Newest receipts

Benchmark rows added to the public ledger for Claude Opus 5 (Adaptive Reasoning, Medium Effort) in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context82%Artificial Analysis4
AA output speedPerformance52 t/sArtificial Analysis4
AA time to first tokenPerformance6.40sArtificial Analysis4
Artificial Analysis Intelligence IndexReasoning45.1%Artificial Analysis4
GPQA DiamondReasoning91.9%Artificial Analysis4
Humanity's Last ExamReasoning51.3%Artificial Analysis4
SciCodeCoding51.5%Artificial Analysis4
τ³-Bench BankingAgentic38.56%Artificial Analysis4
Terminal-BenchAgentic86.14%Artificial Analysis4
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Reasoning92
Coding52
Agentic86
Long context82
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaxAIMiniMaxZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Claude Opus 5 (Adaptive Reasoning, Medium Effort). Scores link to the original source; gaps mean no public row exists yet.

9 sourced rows

9 of 37 catalog benchmarks have a sourced row for Claude Opus 5 (Adaptive Reasoning, Medium Effort).

Reasoning

Coding

Agentic

Long context

Performance

Sources