Which AI model scores highest on HealthBench Professional?

GPT-6 Astra by OpenAI holds the best HealthBench Professional (OpenAI grading, length-adjusted) result among tracked models, at 64.7% (released Sep 3 2026). Higher scores are better on this benchmark.

What are the top 5 models on HealthBench Professional?

1. GPT-6 Astra (OpenAI) — 64.7%; 2. GPT-6 Sol (OpenAI) — 60.8%; 2. GPT-6 Luna (OpenAI) — 60.8%; 4. GPT-5.6 Sol (OpenAI) — 60.5%; 5. GPT-5.6 Terra (OpenAI) — 57.7%.

How many models have a published HealthBench Professional score?

6 tracked models have a published HealthBench Professional (OpenAI grading, length-adjusted) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

Health

HealthBench ProfessionalOpenAI grading, length-adjusted

official OpenAI grading with the response-length penalty applied

HealthBench Professional raw OpenAI grading, raw OpenAI grading, length-adjusted length-adjusted

Rankings

Higher is better

HealthBench Professional — frequently asked questions

: official OpenAI grading with the response-length penalty applied
: GPT-6 Astra by OpenAI holds the best HealthBench Professional (OpenAI grading, length-adjusted) result among tracked models, at 64.7% (released Sep 3 2026). Higher scores are better on this benchmark.
: 1. GPT-6 Astra (OpenAI) — 64.7%; 2. GPT-6 Sol (OpenAI) — 60.8%; 2. GPT-6 Luna (OpenAI) — 60.8%; 4. GPT-5.6 Sol (OpenAI) — 60.5%; 5. GPT-5.6 Terra (OpenAI) — 57.7%.
: 6 tracked models have a published HealthBench Professional (OpenAI grading, length-adjusted) score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.