Ctrl K
Back to models

Sapiens AI

Agnes 2.5 Pro Alpha

  • Released
  • Open weights
  • Apache 2.0
Compare
Intelligence Index

Reasoning setting

26.8
Estimated · Artificial Analysis v4.3.2
Context window
1.05M tok

Performance profile

ReasoningRank #28 of 107
87.6%
GPQA Diamond
CodingRank #54 of 94
42.9%
SciCode

Capabilities

Context window

Modalities

Measured setting

Reasoning trade-offs

Source · archived
Intelligence Index v4.3.2
ReasoningEstimated
Intelligence Index
26.8

Pricing & access

No API offering listed in our tracked sources.

Run locally

Original download
793.6 GB
App memory
Unverified
Documented software
vLLMSetup guide (opens in a new tab)
Find GGUF downloads (opens in a new tab)Original weights (opens in a new tab)

Benchmarks

Full leaderboardSource · archived

Humanity’s Last Exam

Frontier reasoning

Agnes 2.5 Pro Alpha · Reasoning#44 of 126

  1. 1Claude Opus 5.5Max61.4%
  2. 2Claude Fable 5.1Max59.1%
  3. 3Gemini 4 ArgonHigh57.1%
  5. 41Qwen3.8-27BExtra high33.9%
  6. 42Nex-N2-ProReasoning33.7%
  7. 43LongCat 2.0Reasoning33.7%
  8. 44Agnes 2.5 Pro AlphaReasoning33.6%
  9. 45Hy3Reasoning33.5%
  10. 46Inkling SmallReasoning33.3%
  11. 47K2 Horizon 375B A23BReasoning32.0%

Peers use their best measured result on this benchmark.