Back to models
Intelligence Index
Reasoning setting
21.4#51 of 83measured models
Measured · Artificial Analysis v4.3.2- Context window
- 262.14K tok
Performance profile
CodingRank #66 of 94
39.7%
SciCodeCapabilities
Context window
Modalities
Measured setting
Reasoning trade-offs
Source (opens in a new tab)Reasoning
- Intelligence Index
- 21.4
Pricing & access
No API offering listed in our tracked sources.
Run locally
- Original download
- 24.1 GB
- App memory
- Unverified
- Documented software
- vLLM nightlySetup guide (opens in a new tab)
Find GGUF downloadsBenchmarks
Full leaderboardSource (opens in a new tab)24.7%, rank 56 of 12639.7%, rank 66 of 940.0%, rank 56 of 8769.7%, rank 66 of 125979 Elo, rank 53 of 101
Humanity’s Last Exam
Frontier reasoning- 1Claude Opus 5.5Max61.4%
- 2Claude Fable 5.1Max59.1%
- 3Gemini 4 ArgonHigh57.1%
- 53Solar Open2 250BReasoning28.5%
- 54Nemotron 3 Ultra 550B A55BReasoning28.4%
- 55Qwen3.5 122B A10BReasoning25.2%
- 56Nemotron 3.5 Bolt 74B-A9BReasoning24.7%
- 57Gemma 4 31BReasoning23.6%
- 58K2 Horizon MoVA 36B A4BReasoning23.4%
- 59Ling-3.0-flash-FinReasoning22.6%
Peers use their best measured result on this benchmark.
Alternatives
- Scores higherA.X-K2Reasoning
- Intelligence Index
- 21.5+0.1
- More from NVIDIANemotron 3 Ultra 550B A55BReasoning
- Intelligence Index
- 22.9+1.5
- Cost / task
- $0.577
- Similar score
- Intelligence Index
- 22.2+0.8
- Cost / task
- $0.193
Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.