HealthBench Professional: leaderboard
HealthBench professional subset for expert-oriented healthcare question answering.
Metric: Score (%). Source: llm-stats.com. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 | 66 |
| 2 | GPT-6 | 63.4 |
| 3 | GPT-5.6 Sol | 60.5 |
| 4 | Claude Opus 5 | 59.8 |
| 5 | Claude Sonnet 5 | 57.8 |
| 6 | GPT-5.6 Terra | 57.7 |
| 7 | Claude Opus 4.8 | 55.8 |
| 8 | GPT-5.6 Luna | 55.7 |
| 9 | Claude Opus 4.7 | 51.9 |
| 10 | Claude Sonnet 4.6 | 41.7 |
| 11 | GPT-5.5 Instant | 38.4 |
Interactive version: theaggregate.ai/benchmark?slug=healthbench-professional · How It Works · Data refreshed daily, snapshot 2026-09-05.