MoveBench debuts: a rubric-based benchmark for LLM sports and health coaching
Yitong Li · hf · 2026-10-28
MoveBench is a specialized rubric-based benchmark evaluating LLMs on sports and health coaching, covering safety, scientific validity, and personalization across multiple-choice and open-ended tasks. Per its leaderboard, Claude 4.6 Opus leads at 0.943, followed by GPT-5.2 (0.929), GPT-5.5 and Qwen3.5-397B-A17B (0.927 each), while Qwen3-32B trails at 0.830; most models score best on domain adaptation and weakest on QA.
More from Models
- ECI Metric Adopted by Anthropic, Google and Artificial Analysis — Jsevillamol · 2026-10-28
- 'The OpenAI Canyon' meme chart shows OpenAI models trailing on output tokens — steipete · 2026-10-28
- First open-weight production dLLM: Clover-35B hits 680 tps on one GPU, beats same-size AR models on SWE-bench — taoyds · 2026-10-28
- Only Opus 5.5 one-shots "Anthropic" in association test, revealing a persona shift — ryunuck · 2026-10-28
- ChatGPT Pro usage limits cut in half in 3 days, devs voice frustration at OpenAI — johnlindquist · 2026-10-28
- Grok 4.7 scores just 24% on CritPt v2 physics benchmark, trailing Chinese models — teortaxesTex · 2026-10-28