DeepSeek V4.1 Flash 1017 Scores 159 on Epoch ECI, Tied With GPT 5.5
teortaxesTex · x · 2026-11-02
Epoch's Capabilities Index shows open-weights DeepSeek V4.1 Flash 1017 (released Oct 17) scoring 159 ECI — equal to GPT 5.5, +3 over the 0910 build and +10 over original V4-Pro, ranking 19th of 254 tracked models. Notable subscores: ARC-AGI-2 78%, ARC-AGI-1 95%, GPQA Diamond 93%, OTIS Mock AIME 99%, APEX-Agents 79%. The poster expects V4.1-Pro could reach 163 to match 5.6 Sol.
Related event: DeepSeek V4.1 Flash Matches GPT 5.5 on Epoch Capabilities Index(2 posts)→
More from Models
- Phil Chen: the biggest unsolved problem in LLM autoresearch is bypassing Anthropic safeguards — philhchen · 2026-11-02
- Opus 5.5 is good in exactly the ways DeepSeek is good, but not a deep thinker — teortaxesTex · 2026-11-02
- Quant-Arena benchmark: 5 of 9 LLM quant strategies lose money, none beat 17.1% baseline — jiqizhixin · 2026-11-02
- Domingos: If OpenAI and Anthropic Had a Secret Sauce, Distillation Wouldn't Work — pmddomingos · 2026-11-02
- AI commentator laments frontier labs leak nothing, not even model configs — teortaxesTex · 2026-11-02
- First GPT-6.1 Terra test: perfectly rebuilds a scanned PDF form in 81 seconds — MikePFrank · 2026-11-02