Benchmark
PhysicianBench
Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.
- Models measured
- 21
- Scores recorded
- 42
- Sources
- 3
- IQ weight
- 1.69
Score distribution
- 36.3%
- Mean
- 33.6%
- Median
- 18.8%
- Std dev
- 13.7%
- 10th percentile
- 63.2%
- 90th percentile
- 1.4%
- Min
This benchmark has a weight of 1.69 for chat models, measured over 21 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 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 68.4% |
| 2 | Claude Opus 5.5clinical-result-1830· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 68.4% |
| 3 | Claude Fable 5.2· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 64.4% |
| 4 | Claude Sonnet 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 63.2% |
| 5 | Claude Sonnet 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 63.2% |
| 6 | Claude Sonnet 5.5clinical-result-1831· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 63.2% |
| 7 | Claude Fable 5.1adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 61% |
| 8 | Claude Fable 5.1clinical-result-1832· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 61% |
| 9 | Claude Opus 5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 57.6% |
| 10 | Claude Opus 5clinical-result-1833· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 57.6% |
| 11 | Claude Sonnet 5.5clinical-result-1834· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 56.4% |
| 12 | Claude Sonnet 5.5clinical-result-1835· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 47.6% |
| 13 | GPT 5.5clinical-result-167· Clinical Benchmarks data docs | 46.3% |
| 14 | Claude Haiku 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43% |
| 15 | Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 43% |
| 16 | Claude Haiku 5.5clinical-result-2366· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 43% |
| 17 | Claude Sonnet 5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 37.4% |
| 18 | Claude Sonnet 5clinical-result-1836· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 37.4% |
| 19 | Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · xhigh · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 35.8% |
| 20 | Claude Haiku 5.5clinical-result-2367· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 35.8% |
| 21 | GPT-6 Lunaadaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · max · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 33.6% |
| 22 | GPT-6 Lunaclinical-result-2373· Clinical Benchmarks data docs | 33.6% |
| 23 | Claude Opus 4.6clinical-result-168· Clinical Benchmarks data docs | 31.7% |
| 24 | Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · high · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 31.6% |
| 25 | Claude Haiku 5.5clinical-result-2368· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 31.6% |
| 26 | Claude Sonnet 5.5clinical-result-1837· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 30% |
| 27 | Claude Opus 4.7clinical-result-169· Clinical Benchmarks data docs | 29.3% |
| 28 | GPT 5.4clinical-result-170· Clinical Benchmarks data docs | 27.7% |
| 29 | Claude Sonnet 5.5clinical-result-1838· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 27.2% |
| 30 | Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · medium · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 25.2% |
| 31 | Claude Haiku 5.5clinical-result-2369· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 25.2% |
| 32 | Claude Sonnet 4.6clinical-result-171· Clinical Benchmarks data docs | 23% |
| 33 | DeepSeek V4 Proclinical-result-1839· Clinical Benchmarks data docs | 18.7% |
| 34 | Claude Haiku 5.5adaptive thinking · max effort · default sampling · 5-trial mean · context ≤1M unless noted · 100 EHR tasks · low · five attempts/task· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 17.8% |
| 35 | Claude Haiku 5.5clinical-result-2370· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 17.8% |
| 36 | Kimi K2.6clinical-result-172· Clinical Benchmarks data docs | 17% |
| 37 | MiMo V2.5 Proclinical-result-1840· Clinical Benchmarks data docs | 16.7% |
| 38 | Qwen3.6 Plusclinical-result-173· Clinical Benchmarks data docs | 13.7% |
| 39 | MiniMax M2.7clinical-result-1841· Clinical Benchmarks data docs | 8.7% |
| 40 | Gemini 3.1 Pro Previewclinical-result-174· Clinical Benchmarks data docs | 6% |
| 41 | Grok 4.20clinical-result-295· Clinical Benchmarks data docs | 5.3% |
| 42 | Claude Haiku 4.5clinical-result-2372· Clinical Benchmarks data docs· Reported by the model's developer. The source relaying it did not verify it. claim | 1.4% |
Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence) · Data by LLM Stats