Benchmark
HealthBench
What is HealthBench?
Physician-written rubrics scored by an LLM grader, and the length-adjusted variant is what recent cards report. Published by OpenAI.
- Released
- 2025-05-12
- Evaluates
- health
- Openness
- unknown
- Importer
- Model reports
- Reported scores
- 3
Source provenance
- Original evidence https://openai.com/index/healthbench/
HealthBench links
HealthBench results and reported scores
Model reports
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.
Which sources cite HealthBench?
- GPT-5.6 (Sol, Terra, Luna) OpenAI, 2026-07-09
- Seed2.0 (Pro / Lite / Mini) ByteDance, 2026-02-14
- GPT-5 OpenAI, 2025-08-07