HealthBench
HealthBench - healthcare-related benchmark evaluating medical knowledge and reasoning.
Category
General Capabilities
Max Score
—
Score Type
percent
Active
Yes
Model Scores
Scores for HealthBench
| Model | Score | Date | Verified |
|---|---|---|---|
| GPT-5.6 Sol |
72.32%
|
18.08.2026 | Unverified |
| Claude Opus 4.8 |
55.93%
|
18.08.2026 | Unverified |
| Qwen3.7-Max |
67.80%
|
18.08.2026 | Unverified |
| Qwen3.8-2.4T-A95B |
100.00%
|
18.08.2026 | Verified |
| GPT-OSS-20B |
42.50
|
05.08.2025 | Verified |
| GPT-OSS-120B |
85.31%
|
08.08.2025 | Verified |