Claude Sonnet 5 on HealthBench Professional
rank 4 of 9 · updated August 16, 2026
Claude Sonnet 5 scores 0.578 on HealthBench Professional, rank 4 of 9 evaluated models. Anthropic's mid-tier model, positioned for speed at near-Opus quality. HealthBench Professional scores models on 525 tasks drawn from real clinician conversations, graded criterion by criterion against physician-written rubrics on a 0 to 1 scale.
Result and API facts
| rank | 4 of 9 |
|---|---|
| score | 0.578 |
| lab | Anthropic |
| context window | 1.0M tokens |
| API price per 1M tokens | $2.00 in / $10.00 out |
| license | proprietary |
| released | 2026-06-30 |
Position in the field
The gap to the leader, Claude Fable 5 at 0.660, is 0.082. Directly above sits Claude Opus 5 at 0.598. Directly below sits GPT-5.6 Terra at 0.577. Scores on this page come from the same evaluation run, so differences between models are differences on identical tasks, not across configurations.
What does Claude Sonnet 5 score on HealthBench Professional?
Claude Sonnet 5 scores 0.578 on HealthBench Professional, which places it at rank 4 of 9 evaluated models as of August 16, 2026.
How much does Claude Sonnet 5 cost to run?
Claude Sonnet 5 is priced at $2.00 per million input tokens and $10.00 per million output tokens through Anthropic's API.
Head to head
Pairings with a dedicated comparison page are linked; every other difference is in the score-difference matrix.
- Claude Sonnet 5 vs Claude Fable 50.578 vs 0.660 · Claude Fable 5 by 0.082
- Claude Sonnet 5 vs GPT-5.6 Sol0.578 vs 0.605 · GPT-5.6 Sol by 0.027
- Claude Sonnet 5 vs Claude Opus 50.578 vs 0.598 · Claude Opus 5 by 0.020
- 0.578 vs 0.577 · Claude Sonnet 5 by 0.001
- Claude Sonnet 5 vs Claude Opus 4.80.578 vs 0.558 · Claude Sonnet 5 by 0.020
- Claude Sonnet 5 vs GPT-5.6 Luna0.578 vs 0.557 · Claude Sonnet 5 by 0.021
- Claude Sonnet 5 vs GPT-5.5 Instant0.578 vs 0.384 · Claude Sonnet 5 by 0.194
- Claude Sonnet 5 vs MAI-Thinking-10.578 vs 0.350 · Claude Sonnet 5 by 0.228
How tasks are selected and graded is on the methodology page. The full ranking is on the leaderboard.