VerdictPal
Join

Models / Other

Agnes 2.5 Pro Beta

Agnes 2.5 · Released 2026-08-26

AA Intelligence Index 49.1. 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.

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

#27 of 166 on AA Index · current snapshot
49 AA Index · 9 benchmark rows · 1 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite58
ContextUnlisted
Input / 1M$0.10
Output / 1M$0.30
Knowledge cutoffunpublished
Output speed152 t/s
TTFT (default API)1.14s
Statussolid

Family profile

Best published score in each covered benchmark family.

72/100 avg
Reasoning
91
Coding
48
Agentic
70
Long context
78

4 tested benchmark families

Benchmark placements

Where Agnes 2.5 Pro Beta 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#15/ 185Artificial Analysis2026-09-01
AA-LCRLong context#18/ 17978Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#25/ 10636Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#36/ 18449Artificial Analysis2026-09-01
GPQA DiamondReasoning#37/ 18591Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#44/ 18038Artificial Analysis2026-09-01
Terminal-BenchAgentic#45/ 18470Artificial Analysis2026-09-01
SciCodeCoding#57/ 18048Artificial Analysis2026-09-01
AA time to first tokenPerformance#59/ 185Artificial Analysis2026-09-01
Family context

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

ReasoningAgnes 2.5 Pro Beta · 91
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingAgnes 2.5 Pro Beta · 48
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticAgnes 2.5 Pro Beta · 70
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextAgnes 2.5 Pro Beta · 78
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Newest receipts

Benchmark rows added to the public ledger for Agnes 2.5 Pro Beta in the last 120 days. Older rows live in the full table below.

9 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context78%Artificial Analysis2
AA output speedPerformance152 t/sArtificial Analysis2
AA time to first tokenPerformance1.14sArtificial Analysis2
Artificial Analysis Intelligence IndexReasoning49.1%Artificial Analysis2
GPQA DiamondReasoning90.5%Artificial Analysis2
Humanity's Last ExamReasoning37.5%Artificial Analysis2
SciCodeCoding47.6%Artificial Analysis2
τ³-Bench BankingAgentic35.67%Artificial Analysis2
Terminal-BenchAgentic69.66%Artificial Analysis2
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.

Reasoning91
Coding48
Agentic70
Long context78
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover or tab any dot for name, score, and input price.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for Agnes 2.5 Pro Beta. 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 Beta.

Reasoning

Coding

Agentic

Long context

Performance

Sources