HealthBench: leaderboard
HealthBench: Evaluates clinical, biomedical, medical-exam, coding, or healthcare-document reasoning.
Metric: Score (%). Source: openai.com. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 | 60 | #124 |
| 2 | GPT-4.1 | 48 | #248 |
| 3 | Grok 3 | 46 | #304 |
| 4 | Gemini 2.5 Pro | 44 | #152 |
| 5 | O1 | 42 | #183 |
| 6 | GPT-4.1 Nano | 34 | #731 |
| 7 | GPT-4o | 32 | #343 |
| 8 | Claude 3.7 Sonnet | 30 | #249 |
| 9 | Llama 4 Maverick | 26 | #464 |
| 10 | GPT-3.5 Turbo | 16 | #865 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=healthbench · How It Works · Data refreshed daily, snapshot 2026-10-20.