Benchmark
HealthAgentBench
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 10
- Scores recorded
- 12
- Sources
- 1
- IQ weight
- 1.18
Score distribution
- 31.2%
- Mean
- 32%
- Median
- 11.6%
- Std dev
- 17%
- 10th percentile
- 45%
- 90th percentile
- 16%
- Min
This benchmark has a weight of 1.18 for chat models, measured over 10 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5clinical-result-129· Clinical Benchmarks data docs | 55% |
| 2 | GPT-5.6 Solclinical-result-130· Clinical Benchmarks data docs | 45% |
| 3 | GPT 5.5clinical-result-131· Clinical Benchmarks data docs | 42% |
| 4 | Claude Opus 4.8clinical-result-132· Clinical Benchmarks data docs | 36% |
| 5 | GPT 5.5clinical-result-133· Clinical Benchmarks data docs | 35% |
| 6 | Claude Opus 4.8clinical-result-134· Clinical Benchmarks data docs | 32% |
| 7 | GPT 5.4clinical-result-321· Clinical Benchmarks data docs | 28% |
| 8 | Claude Opus 4.7clinical-result-322· Clinical Benchmarks data docs | 27% |
| 9 | GPT-5.3 Codexclinical-result-1822· Clinical Benchmarks data docs | 22% |
| 10 | Claude Opus 4.6clinical-result-323· Clinical Benchmarks data docs | 19% |
| 11 | Claude Sonnet 4.6clinical-result-324· Clinical Benchmarks data docs | 17% |
| 12 | GPT-5.4 miniclinical-result-325· Clinical Benchmarks data docs | 16% |
Data credits: Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence)