VerdictPal
Join

Models / Other

Agnes 2.5 Pro Alpha

Agnes 2.5 · Released 2026-07-24

AA Intelligence Index 38.8. 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.

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

#68 of 184 on AA Index · current snapshot
27 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite49
Context128K
Input / 1M$0.45
Output / 1M$0.90
Knowledge cutoffopen
Output speed196 t/s
TTFT (default API)1.73s
Statussolid

Family profile

Best published score in each covered benchmark family.

69/100 avg
Reasoning
88
Coding
43
Agentic
67
Long context
76

4 tested benchmark families

Benchmark placements

Where Agnes 2.5 Pro Alpha 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
AA output speedPerformance#11/ 190Artificial Analysis2026-09-10
AA time to first tokenPerformance#40/ 190Artificial Analysis2026-09-10
Terminal-BenchAgentic#57/ 18967Artificial Analysis2026-09-10
Humanity's Last ExamReasoning#64/ 18534Artificial Analysis2026-09-10
SciCodeCoding#68/ 9443Artificial Analysis2026-09-10
GPQA DiamondReasoning#70/ 19088Artificial Analysis2026-09-10
Artificial Analysis Intelligence IndexReasoning#71/ 18827Artificial Analysis2026-09-10
AA-LCRLong context#78/ 18276Artificial Analysis2026-09-10
τ³-Bench BankingAgentic#87/ 11112Artificial Analysis2026-09-10
Family context

The three highest-scoring models with pages in each capability family. Where Agnes 2.5 Pro Alpha shows up, it's highlighted.

ReasoningAgnes 2.5 Pro Alpha · 88
  1. 1
    GPT-6 Astra
    96
  2. 2
    Claude Mythos Preview
    95
  3. 3
    Gemini 3.7 Flash
    95
CodingAgnes 2.5 Pro Alpha · 43
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticAgnes 2.5 Pro Alpha · 67
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextAgnes 2.5 Pro Alpha · 76
  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 Agnes 2.5 Pro Alpha in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context75.67%Artificial Analysis4
AA output speedPerformance196 t/sArtificial Analysis4
AA time to first tokenPerformance1.73sArtificial Analysis4
Artificial Analysis Intelligence IndexReasoning26.8%Artificial Analysis4
GPQA DiamondReasoning87.6%Artificial Analysis4
Humanity's Last ExamReasoning33.6%Artificial Analysis4
SciCodeCoding42.9%Artificial Analysis4
τ³-Bench BankingAgentic12.37%Artificial Analysis4
Terminal-BenchAgentic67.04%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.

Reasoning88
Coding43
Agentic67
Long context76
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 Agnes 2.5 Pro Alpha. 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 Agnes 2.5 Pro Alpha.

Reasoning

Coding

Agentic

Long context

Performance

Sources