DeepSeek V4.1F-1017 Hits 159 on Epoch Index, Matching GPT 5.5, 10 Points from Astra
teortaxesTex · x · 2026-11-02
teortaxesTex analyzes Epoch Capabilities Index scores and argues the gap to the true frontier is underrated.
Key numbers:
- DeepSeek V4.1F-1017 scores 159, equal to GPT 5.5 — 3 points over V4.1F-0910 and 10 over the original V4-Pro
- Astra is another 10 points ahead; at 163, a model would match GPT 5.6 Sol
- With Astra having saturated many evals and even V4.1 close behind, the author argues the Epoch index needs a major refresh and hopes Lisan speeds that up
He hopes DeepSeek V4.1-Pro at least reaches the 163-point level.
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
- DeepSeek V4.1 Flash 1017 Scores 159 on Epoch ECI, Tied With GPT 5.5 — teortaxesTex · 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