Claude Opus 5: healthcare benchmark results
Anthropic · 7 boards · updated August 16, 2026
The index currently holds 7 results for Claude Opus 5: 0.598 on HealthBench Professional (3 of 9), 67.1 on HealthBench (1 of 8), 78.3 on Health Optimization Bench (3 of 15), 57.1% on MAST (Medical AI Superintelligence Test) (6 of 11), 74.6% on First, Do NOHARM (v2) (1 of 19), 63.57% on MedCode (Vals AI) (1 of 85), 90.99% on MedScribe (Vals AI) (1 of 84). It tops HealthBench and First, Do NOHARM (v2) and MedCode (Vals AI) and MedScribe (Vals AI). Scores below sit on different scales and come from different graders, so read each against its own benchmark, never against the others.
Results by benchmark
| benchmark | score | position | as of |
|---|---|---|---|
| HealthBench Professional via healthbenchprofessional.com | 0.598 | 3 of 9 | 2026-08 |
| HealthBench via OpenAI Deployment Safety Hub (GPT-5.6 system card + August 2026 updates); benchlm.ai and llm-stats.com mirror | 67.1 | 1 of 8 | 2026-08 |
| Health Optimization Bench via healthoptimizationbench.com | 78.3 | 3 of 15 | 2026-08 |
| MAST (Medical AI Superintelligence Test) via ARISE MAST leaderboard | 57.1% | 6 of 11 | 2026-08 |
| First, Do NOHARM (v2) via ARISE MAST technical leaderboard | 74.6% | 1 of 19 | 2026-08 |
| MedCode (Vals AI) via Vals AI MedCode leaderboard | 63.57% | 1 of 85 | 2026-08 |
| MedScribe (Vals AI) via Vals AI MedScribe leaderboard | 90.99% | 1 of 84 | 2026-08 |
Position counts against the source's full board, including rows this index does not mirror. Config caveats, where a source noted any, are on each benchmark's page.
Which healthcare benchmarks is Claude Opus 5 scored on?
As of August 16, 2026, Claude Opus 5 holds current results on 7 tracked benchmarks: HealthBench Professional, HealthBench, Health Optimization Bench, MAST (Medical AI Superintelligence Test), First, Do NOHARM (v2), MedCode (Vals AI), MedScribe (Vals AI).
How does Claude Opus 5 rank on them?
Claude Opus 5 stands at 0.598 on HealthBench Professional (3 of 9), 67.1 on HealthBench (1 of 8), 78.3 on Health Optimization Bench (3 of 15), 57.1% on MAST (Medical AI Superintelligence Test) (6 of 11), 74.6% on First, Do NOHARM (v2) (1 of 19), 63.57% on MedCode (Vals AI) (1 of 85), 90.99% on MedScribe (Vals AI) (1 of 84). It holds first place on HealthBench and First, Do NOHARM (v2) and MedCode (Vals AI) and MedScribe (Vals AI).
The benchmarks themselves are described on their pages, linked in the table above, and the whole field is on the index.