GPT-6 Astra scores 87.4% on MedXpertQA, beating every other LLM
toshv · reddit · 2026-09-11
A Reddit user ran GPT-6 Astra on MedXpertQA, an expert-level medical knowledge benchmark, scoring 87.4% — above every other model tested from OpenAI and Google. Two years ago GPT-4o scored just 42.8%; Astra is approaching saturation. Full results are filterable by organ system at natomy.com/med-evals, and the author takes requests for other models.
More from Models
- Kimi K3.1 Pro tops MathArena open models, closing frontier gap to about 6 months — teortaxesTex · 2026-10-26
- Ruff creator Charlie Marsh: frontier models find infinite bugs in any software — charliermarsh · 2026-10-26
- Dev testing one skill across Opus 5.5 and Sol 6.1 finds instruction following 'all over the map' — lucasmeijer · 2026-10-26
- supermemory unveils learner-1, claiming continual learning for any model — julianweisser · 2026-10-26
- Anthropic says Opus 5.5 cuts cost per task by ~40% vs Opus 5 — taufiqintech · 2026-10-26
- AI search APIs ranked: top 12 within 7 points, yet prices differ nearly 7x — FinanceYF5 · 2026-10-26