Updated daily at 05:00 UTC
Benchmark
HealthBench Professional
Measures: Models respond to 525 real clinician conversations (care consult, writing and documentation, medical research; about one-third adversarial), graded against physician-written, multi-physician-adjudicated rubrics; score = earned/possible rubric points, clipped to 0-1.
- Maintainer
- OpenAI
- Introduced
- 2026
- Status
- Active
- Saturation
- Headroom
- Typical provenance
- Self-reported
- In scoring set
- No
Leaderboard
Independent and third-party results only.
No independent or third-party results yet. Show all results