GPT-5.6 Sol vs Claude Opus 5 on HealthBench Professional
updated August 16, 2026
GPT-5.6 Sol scores 0.605 to Claude Opus 5's 0.598, a gap of 0.007 on the 525 physician-graded tasks of HealthBench Professional. The table below puts the scores next to what each model costs to actually run.
Side by side
 GPT-5.6 Sol |  Claude Opus 5 | |
|---|---|---|
| score | 0.605 | 0.598 |
| rank | 2 of 9 | 3 of 9 |
| context window | 1.1M | 1.0M |
| price per 1M tokens, in / out | $5.00 / $30.00 | $5.00 / $25.00 |
| 1,000 consult exchanges | $31.00 | $27.50 |
| released | 2026-07-09 | 2026-07-24 |
| license | proprietary | proprietary |
Consult exchange: 2,000 input and 700 output tokens, priced at list rates as of August 16, 2026. GPT-5.6 models charge higher rates above 272K input tokens. MAI-Thinking-1 is in public preview on Microsoft Foundry without final list pricing.
Reading this pairing
Priced identically at $5 per million input tokens, these two are the natural cross-lab comparison, and the score difference is 0.007 on a 525-task set, small enough that a re-run could plausibly flip it. Sol charges $5 more per million output tokens and offers a slightly larger context window. On this benchmark alone there is no winner here; the honest tiebreakers are output price, ecosystem, and how each model behaves in your own scaffolding.
Which scores higher on HealthBench Professional, GPT-5.6 Sol or Claude Opus 5?
GPT-5.6 Sol scores higher: 0.605 against Claude Opus 5's 0.598, a difference of 0.007 on the 525-task set, as of August 16, 2026.
Which is cheaper to run, GPT-5.6 Sol or Claude Opus 5?
Claude Opus 5. 1,000 typical consult exchanges (2,000 input and 700 output tokens each) cost $27.50 against $31.00 at list rates.
Related comparisons
- 0.660 vs 0.605
- 0.660 vs 0.598
- 0.598 vs 0.558
Full results for both models: GPT-5.6 Sol and Claude Opus 5. The complete score-difference matrix is on the compare page.