Claude Sonnet 5.5 and GPT-6 Astra tie atop updated medical record review leaderboard at 41.7%
ArtificialAnlys · x · 2026-10-21
Artificial Analysis overhauled MLCR-AA, its leaderboard for Wisedocs' Medical Long Context Reasoning benchmark: models are now prompted as medical record reviewers answering in one paragraph with conclusions and cited evidence, graded by a three-model judge panel, with all 60 expert reference answers re-reviewed.
Key results:
- Claude Sonnet 5.5 (max with fallback) and GPT-6 Astra (max) tie for #1 at 41.7%, followed by GPT-6.1 Sol (max) at 38.3%; Qwen3.8 2.4T A95B leads open weights at 30.6%.
- Medical record review remains far from solved: no model passes more than 42% of questions and the median model passes under 15%.
- Anthropic models sweep the top 18 completeness scores (best: 64.8%); OpenAI models sweep the top 27 accuracy scores (GPT-6.1 Sol max at 96.7% accuracy but only 40% complete).
More from Models
- Recursive AI ships Flint, a 40B-A2.5B MoE claimed to match Qwen3.6-27B, weights on Hugging Face — ChengleiSi · 2026-10-22
- DeepSeek ships V4.1-Flash-1017, now the #1 open model on RSI-Inference Bench — scaling01 · 2026-10-22
- OpenAI's Cybersecurity Guardrails Block a Dev From Auditing His Own Patched Server — Next_Technology6361 · 2026-10-22
- TPN-006 hits Hugging Face: 7.7GB compressed model beats its 17GB Mistral teacher — markjeffrey · 2026-10-22
- Kimi K3.1's latest checkpoint draws 49k-token SVG portrait zero-shot in 22 minutes — lyraxana · 2026-10-22
- MiniMax launches M3.1 Flash at 1/3 the cost and 3x the speed of M3, weights out Oct 23 — MiniMax_AI · 2026-10-22