DeepSeek V4.1 Flash 1017 ranks 5th on new Noahbench eval, "cheap, slow, decent"
teortaxesTex · x · 2026-10-18
New Noahbench results place DeepSeek V4.1 Flash 1017 in 5th, described as "cheap, slow, and pretty decent." teortaxesTex suggests the model's slow runtime may stem from the same rabbitholing problem affecting other models.
More from Models
- AutoTrust's HF models hit top trending with 5.7K likes in 17 days — accounts look farmed — realmrfakename · 2026-10-18
- Google AI Studio Build upgraded with Gemini 4 Argon and one-click Cloud Run deploys — clmt · 2026-10-18
- Rumor: OpenAI to reveal new model next week, 'aurora' alpha variants spotted — imjustnewatai · 2026-10-18
- User reports Gemini hanging and slow since Argon launch, hardware said to be diverted — johnfromberkeley · 2026-10-18
- Domingos: superposition is a feature, not a bug — it boosts memory, not blocks interpretability — pmddomingos · 2026-10-18
- Step 5 Preview open-sourced after ranking 2nd by usage on OpenRouter this week — Xianbao_QIAN · 2026-10-18