Back to models
Intelligence Index
Max setting
34.8#29 of 83measured models
Measured · Artificial Analysis v4.3.2- Context window
- 1M tok
Performance profile
ReasoningRank #16 of 107
91.3%
GPQA DiamondCodingRank #33 of 94
49.7%
SciCodeCost efficiencyRank #27 of 67
110.8pts / $
Intelligence ÷ eval costCapabilities
Context window
Modalities
Measured setting
Reasoning trade-offs
Source (opens in a new tab)DeepSeek V4 Flash Vision
Cost per task (USD) · log scale
DeepSeek V4 Flash VisionMax
- Intelligence Index
- 34.8
- Cost / task
- $0.3144
- Evaluation time
- 4.7m
Pricing & access
No API offering listed in our tracked sources.
Benchmarks
Full leaderboardSource (opens in a new tab)34.5%, rank 39 of 12649.7%, rank 33 of 9412.1%, rank 28 of 8781.3%, rank 23 of 1251,547 Elo, rank 23 of 101
Humanity’s Last Exam
Frontier reasoning- 1Claude Opus 5.5Max61.4%
- 2Claude Fable 5.1Max59.1%
- 3Gemini 4 ArgonHigh57.1%
- 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%
- 42Nex-N2-ProReasoning33.7%
Peers use their best measured result on this benchmark.
Alternatives
- Scores higher
- Intelligence Index
- 36.0+1.2
- Cost / task
- $0.674+114%
- Higher score, lower costMiMo-V2.6-ProReasoning
- Intelligence Index
- 46.3+11.5
- Cost / task
- $0.133−58%
- More from DeepSeek
- Intelligence Index
- 39.5+4.6
- Cost / task
- $0.265−16%
Each model at its highest-scoring measured setting · Intelligence Index v4.3.2 and its evaluation cost per task.