Claude Opus 5.5 on HealthBench Professional
rank 4 of 31 · updated September 30, 2026
Claude Opus 5.5 scores 0.656 on HealthBench Professional, rank 4 of 31 models on the board. 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 31 |
|---|---|
| score | 0.656 |
| lab | Anthropic |
| context window | 1M tokens |
| API price per 1M tokens | $4.00 in / $20.00 out |
| license | proprietary |
| source | Claude Opus 5.5 System Card (system card) |
| released | 2026-09-22 |
Position in the field
The gap to the leader, GPT-6 Astra (Anthropic run) at 0.703, is 0.047. Directly above sits Claude Fable 5 at 0.660. Directly below sits GPT-6 Astra at 0.647. The scores on this page are compiled from published documents rather than from one controlled run, so a small gap between 2 models can reflect a difference in grader version, reasoning effort, or deployment setting as well as a difference in capability.
Source of this score
Read from Claude Opus 5.5 System Card (system card, Anthropic, 2026-09-22). Vendor-reported. Confidence: verified. Configuration: Anthropic; adaptive max effort; Opus 4.8 grader; five trials; no tools or custom system prompt; safety classifiers and refusal fallback to Opus 5; length-adjusted; raw 77.1%.
8.15.2 HealthBench Professional results length adjustment, which penalizes verbose model responses, Claude Opus 5.5 achieved a score of 65.6%.
pp. 213–214, Section 8.15.2, paragraph and Figure 8.15.2.A. · full entry on the sources page
What does Claude Opus 5.5 score on HealthBench Professional?
Claude Opus 5.5 scores 0.656 on HealthBench Professional, which places it at rank 4 of 31 models on the board as of September 30, 2026. The number was read from Claude Opus 5.5 System Card, listed on the sources page.
How much does Claude Opus 5.5 cost to run?
Claude Opus 5.5 is priced at $4.00 per million input tokens and $20.00 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 Opus 5.5 vs GPT-6 Astra (Anthropic run)0.656 vs 0.703 · GPT-6 Astra (Anthropic run) by 0.047
- Claude Opus 5.5 vs Claude Sonnet 5.50.656 vs 0.692 · Claude Sonnet 5.5 by 0.036
- Claude Opus 5.5 vs Claude Fable 50.656 vs 0.660 · Claude Fable 5 by 0.004
- Claude Opus 5.5 vs GPT-6 Astra0.656 vs 0.647 · Claude Opus 5.5 by 0.009
- Claude Opus 5.5 vs Claude Fable 5 (September card)0.656 vs 0.633 · Claude Opus 5.5 by 0.023
- Claude Opus 5.5 vs Claude Fable 5.10.656 vs 0.621 · Claude Opus 5.5 by 0.035
- Claude Opus 5.5 vs GPT-6 Luna0.656 vs 0.608 · Claude Opus 5.5 by 0.048
- Claude Opus 5.5 vs GPT-6 Sol0.656 vs 0.608 · Claude Opus 5.5 by 0.048
- Claude Opus 5.5 vs GPT-5.6 Sol0.656 vs 0.605 · Claude Opus 5.5 by 0.051
- Claude Opus 5.5 vs Claude Opus 50.656 vs 0.598 · Claude Opus 5.5 by 0.058
- Claude Opus 5.5 vs Muse Spark 1.10.656 vs 0.593 · Claude Opus 5.5 by 0.063
- Claude Opus 5.5 vs Claude Sonnet 50.656 vs 0.578 · Claude Opus 5.5 by 0.078
- Claude Opus 5.5 vs GPT-5.6 Terra0.656 vs 0.577 · Claude Opus 5.5 by 0.079
- Claude Opus 5.5 vs Claude Opus 4.8 (Opus 4.8 grader)0.656 vs 0.574 · Claude Opus 5.5 by 0.082
- Claude Opus 5.5 vs Grok 4.70.656 vs 0.567 · Claude Opus 5.5 by 0.089
- Claude Opus 5.5 vs Claude Opus 4.80.656 vs 0.558 · Claude Opus 5.5 by 0.098
- Claude Opus 5.5 vs GPT-5.6 Luna0.656 vs 0.557 · Claude Opus 5.5 by 0.099
- Claude Opus 5.5 vs Muse Spark0.656 vs 0.541 · Claude Opus 5.5 by 0.115
- Claude Opus 5.5 vs GPT-5.6 Sol (August)0.656 vs 0.540 · Claude Opus 5.5 by 0.116
- Claude Opus 5.5 vs Claude Opus 4.70.656 vs 0.519 · Claude Opus 5.5 by 0.137
- Claude Opus 5.5 vs GPT-5.50.656 vs 0.518 · Claude Opus 5.5 by 0.138
- Claude Opus 5.5 vs Grok 4.60.656 vs 0.485 · Claude Opus 5.5 by 0.171
- Claude Opus 5.5 vs GPT-5.40.656 vs 0.481 · Claude Opus 5.5 by 0.175
- Claude Opus 5.5 vs GPT-50.656 vs 0.462 · Claude Opus 5.5 by 0.194
- Claude Opus 5.5 vs GPT-5.20.656 vs 0.459 · Claude Opus 5.5 by 0.197
- Claude Opus 5.5 vs Claude Sonnet 4.60.656 vs 0.442 · Claude Opus 5.5 by 0.214
- Claude Opus 5.5 vs GPT-5.6 Luna (August)0.656 vs 0.441 · Claude Opus 5.5 by 0.215
- Claude Opus 5.5 vs GPT-5.10.656 vs 0.396 · Claude Opus 5.5 by 0.260
- Claude Opus 5.5 vs GPT-5.5 Instant0.656 vs 0.384 · Claude Opus 5.5 by 0.272
- Claude Opus 5.5 vs MAI-Thinking-10.656 vs 0.350 · Claude Opus 5.5 by 0.306
Where the scores come from and how they are read is on the methodology page. The full ranking is on the leaderboard.