Skip to content

Benchmark

Vals MedScribe

Scores are percentages. Higher is better.

Models measured
103
Scores recorded
207
Sources
2
IQ weight
2.20

Score distribution

78.6%
Mean
80.8%
Median
11.3%
Std dev
67.7%
10th percentile
88.5%
90th percentile
4.3%
Min
4.3%91.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.20 for chat models, measured over 103 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51DeepSeek V4.1 Flash· Vals AI85.5%
52DeepSeek V4.1 Flashclinical-result-2162· Clinical Benchmarks data docs85.5%
53Inkling FP4clinical-result-2163· Clinical Benchmarks data docs85.4%
54Inkling FP4· Vals AI85.4%
55Claude Opus 4.5· Vals AI85.3%
56Claude Opus 4.5clinical-result-2164· Clinical Benchmarks data docs85.3%
57MiMo V2.6 Flashclinical-result-2165· Clinical Benchmarks data docs85.3%
58MiMo V2.6 Flash· Vals AI85.3%
59GPT-5.6 Sol· Vals AI85.2%
60Claude Haiku 4.5clinical-result-2167· Clinical Benchmarks data docs85.2%
61Claude Haiku 4.5thinking· Vals AI85.2%
62GPT-5.6 Solclinical-result-2166· Clinical Benchmarks data docs85.2%
63Qwen3.8 Maxclinical-result-2168· Clinical Benchmarks data docs85%
64Qwen3.8 Max· Vals AI84.9%
65Claude Sonnet 4.5clinical-result-2169· Clinical Benchmarks data docs84.5%
66Claude Sonnet 4.5· Vals AI84.5%
67Gemini 3.8 Flashclinical-result-2170· Clinical Benchmarks data docs84.5%
68Gemini 3.8 Flash· Vals AI84.5%
69GPT-5.6 Luna· Vals AI84.4%
70GPT 5.2clinical-result-2172· Clinical Benchmarks data docs84.4%
71GPT-5.6 Lunaclinical-result-2171· Clinical Benchmarks data docs84.4%
72GPT 5.2· Vals AI84.4%
73Inkling Small· Vals AI84.1%
74Inkling Smallclinical-result-2173· Clinical Benchmarks data docs84.1%
75Claude Sonnet 4.5clinical-result-2174· Clinical Benchmarks data docs84.1%
76Gemini 3.7 Flash· Vals AI83.9%
77Gemini 3.7 Flashclinical-result-2175· Clinical Benchmarks data docs83.9%
78Qwen 3.8 27Bclinical-result-2176· Clinical Benchmarks data docs83.8%
79Qwen 3.8 27B· Vals AI83.8%
80MiMo V2.5 Pro· Vals AI83.7%
81MiMo V2.5 Proclinical-result-2177· Clinical Benchmarks data docs83.7%
82GPT-6 Luna· Vals AI83.7%
83GPT-6 Lunaclinical-result-2178· Clinical Benchmarks data docs83.7%
84GPT 5· Vals AI83.7%
85GPT 5clinical-result-2179· Clinical Benchmarks data docs83.7%
86Hy4 preview· Vals AI83.6%
87Hy4 previewclinical-result-2180· Clinical Benchmarks data docs83.6%
88GLM 5.2· Vals AI83.5%
89GLM 5.2clinical-result-2181· Clinical Benchmarks data docs83.5%
90Claude Opus 4.5clinical-result-2182· Clinical Benchmarks data docs83.3%
91Gemini 2.5 Flash· Vals AI83%
92Gemini 2.5 Flashclinical-result-2183· Clinical Benchmarks data docs83%
93Claude Opus 4.7· Vals AI83%
94Claude Opus 4.7clinical-result-2184· Clinical Benchmarks data docs83%
95Gemini 2.5 Flashclinical-result-2185· Clinical Benchmarks data docs82.9%
96GPT-5.6 Terraclinical-result-2186· Clinical Benchmarks data docs82.9%
97GPT-5.6 Terra· Vals AI82.9%
98GPT-6 Sol· Vals AI82%
99GPT-6 Solclinical-result-2187· Clinical Benchmarks data docs82%
100Grok 4 Fast· Vals AI81.6%

Showing 51–100 of 207 results.

Data credits: Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence)