Back to models
Intelligence Index
Reasoning setting
30.3#33 of 83measured models
Measured · Artificial Analysis v4.3.2- API price from
- $0.25 in / $1.00 out / 1M tokOpenRouter · Nebius · FP8
- Context window
- 1.05M tok
Performance profile
CodingRank #37 of 94
48.5%
SciCodeCost efficiencyRank #35 of 67
53.6pts / $
Intelligence ÷ eval costCapabilities
Context window
Modalities
Measured setting
Integration
Reasoning trade-offs
Source (opens in a new tab)Beam 1.0
Cost per task (USD) · log scale
Beam 1.0Reasoning
- Intelligence Index
- 30.3
- Cost / task
- $0.5657
- Evaluation time
- 1.4m
Pricing & access
USD per 1M tokens
- Input
- $1.30
- Output
- $4.30
- Cached input
- $0.13
Run locally
- Original download
- 1,002.7 GB
- App memory
- Unverified
- Documented software
- vLLM / SGLangSetup guide (opens in a new tab)
Find GGUF downloadsBenchmarks
Full leaderboardSource (opens in a new tab)34.7%, rank 38 of 12648.5%, rank 37 of 943.0%, rank 34 of 8778.0%, rank 44 of 1251,177 Elo, rank 39 of 101
Humanity’s Last Exam
Frontier reasoning- 1Claude Opus 5.5Max61.4%
- 2Claude Fable 5.1Max59.1%
- 3Gemini 4 ArgonHigh57.1%
- 35MiMo-V2.6-FlashReasoning35.1%
- 36Mistral Large 4Reasoning35.0%
- 36Kimi K2.7 CodeReasoning35.0%
- 38Beam 1.0Reasoning34.7%
- 39DeepSeek V4 Flash VisionMax34.5%
- 40Apodex 1.1Reasoning34.1%
- 41Qwen3.8-27BExtra high33.9%
Peers use their best measured result on this benchmark.
Alternatives
- Scores higherK2 Horizon 375B A23BReasoning
- Intelligence Index
- 30.5+0.2
- Higher score, lower costMiMo-V2.6-ProReasoning
- Intelligence Index
- 46.3+16.0
- Cost / task
- $0.133−76%
- Similar scoreGemini 3.1 Pro PreviewReasoning
- Intelligence Index
- 29.7−0.6
- Cost / task
- $1.30+130%
Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.