GPT-6 Astra: healthcare benchmark results
OpenAI · 6 boards · updated September 8, 2026
The index currently holds 6 results for GPT-6 Astra: 0.634 on HealthBench Professional (2 of 22 indexed rows), 0.363 on HealthBench Hard (2 of 16 indexed rows), 58.1 on HealthBench (9 of 18 indexed rows), 48.49% on MedCode (Vals AI) (11 of 11 indexed rows), 87.91% on MedScribe (Vals AI) (8 of 12 indexed rows), 1.000 on OpenAI Dynamic Mental Health Evaluations (1 of 15 indexed rows). It tops OpenAI Dynamic Mental Health Evaluations. Scores below sit on different scales and come from different graders, so read each against its own benchmark, never against the others.
Results by benchmark
| benchmark | score | index position | as of |
|---|---|---|---|
| HealthBench Professional | 0.634 | 2 of 22 | 2026-09 |
| HealthBench Hard | 0.363 | 2 of 16 | 2026-09 |
| HealthBench | 58.1 | 9 of 18 | 2026-09 |
| MedCode (Vals AI) | 48.49% | 11 of 11 source board: 85 | 2026-09 |
| MedScribe (Vals AI) | 87.91% | 8 of 12 source board: 84 | 2026-09 |
| OpenAI Dynamic Mental Health Evaluations | 1.000 | 1 of 15 | 2026-09 |
Position follows the rows held in this index, including separate configurations. A source board may contain additional rows; its size is listed separately when it differs. Config caveats, where a source noted any, are on each benchmark's page, and the document behind each score is cited on the sources page.
Which healthcare benchmarks is GPT-6 Astra scored on?
As of September 8, 2026, GPT-6 Astra holds current results on 6 tracked benchmarks: HealthBench Professional, HealthBench Hard, HealthBench, MedCode (Vals AI), MedScribe (Vals AI), OpenAI Dynamic Mental Health Evaluations.
How does GPT-6 Astra rank on them?
GPT-6 Astra stands at 0.634 on HealthBench Professional (2 of 22 indexed rows), 0.363 on HealthBench Hard (2 of 16 indexed rows), 58.1 on HealthBench (9 of 18 indexed rows), 48.49% on MedCode (Vals AI) (11 of 11 indexed rows), 87.91% on MedScribe (Vals AI) (8 of 12 indexed rows), 1.000 on OpenAI Dynamic Mental Health Evaluations (1 of 15 indexed rows). It holds first place on OpenAI Dynamic Mental Health Evaluations.
The benchmarks themselves are described on their pages, linked in the table above, and the whole field is on the index.