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
101Grok 4 Fastclinical-result-2188, reasoning· Clinical Benchmarks data docs81.6%
102Ling 3.0 Flashclinical-result-2189· Clinical Benchmarks data docs80.9%
103MiniMax M2.1clinical-result-2190· Clinical Benchmarks data docs80.8%
104MiniMax M2.1· Vals AI80.8%
105GPT-5 miniclinical-result-2191· Clinical Benchmarks data docs80.6%
106GPT-5 mini· Vals AI80.6%
107Mistral Large 4· Vals AI80.4%
108DeepSeek V4 Flash 0731· Vals AI80.4%
109DeepSeek V4 Flashclinical-result-2192· Clinical Benchmarks data docs80.4%
110DeepSeek V4 Pro 0813· Vals AI80.2%
111DeepSeek V4 Proclinical-result-2193· Clinical Benchmarks data docs80.2%
112MiniMax M2.7clinical-result-2194· Clinical Benchmarks data docs79.9%
113MiniMax M2.7· Vals AI79.9%
114Grok 4 Fastclinical-result-2195, non-reasoning· Clinical Benchmarks data docs79.7%
115Gemini 3.6 Flash· Vals AI79.7%
116Gemini 3.6 Flashclinical-result-2196· Clinical Benchmarks data docs79.7%
117Qwen3.7 Maxclinical-result-2197· Clinical Benchmarks data docs79.4%
118Qwen3.7 Max· Vals AI79.4%
119Grok 4.1 Fast· Vals AI78.7%
120Grok 4.1 Fastclinical-result-2198, reasoning· Clinical Benchmarks data docs78.7%
121Gemini 2.5 Flashclinical-result-2199· Clinical Benchmarks data docs78.5%
122gemini-2.5-flash-preview-09-2025thinking· Vals AI78.5%
123Grok 4· Vals AI78.2%
124Grok 4clinical-result-2200· Clinical Benchmarks data docs78.2%
125Kimi K2.6clinical-result-2201· Clinical Benchmarks data docs78.2%
126Kimi K2.6· Vals AI78.1%
127Gemini 2.5 Flashclinical-result-2202· Clinical Benchmarks data docs78%
128gemini-2.5-flash-preview-09-2025· Vals AI77.9%
129GPT 5.4clinical-result-2203· Clinical Benchmarks data docs77.6%
130GPT 5.4· Vals AI77.5%
131Grok 4.1 Fastclinical-result-2204, non-reasoning· Clinical Benchmarks data docs77.5%
132Qwen3 VL Plusclinical-result-2205· Clinical Benchmarks data docs77.1%
133Qwen3 VL Plus· Vals AI77.1%
134GPT 5.4 Nano· Vals AI77.1%
135GPT 5.4 Nanoclinical-result-2206· Clinical Benchmarks data docs77.1%
136Qwen3.6 Plus· Vals AI77%
137Qwen3.6 Plusclinical-result-2207· Clinical Benchmarks data docs77%
138OpenAI o3· Vals AI76.7%
139OpenAI o3clinical-result-2208· Clinical Benchmarks data docs76.7%
140Gemini 3.5 Flash· Vals AI76.6%
141Gemini 3.5 Flashclinical-result-2209· Clinical Benchmarks data docs76.6%
142Kimi K2.5thinking· Vals AI76.4%
143Kimi K2.5clinical-result-2210· Clinical Benchmarks data docs76.4%
144Gemini 3.1 Pro Preview· Vals AI76.1%
145Gemini 3.1 Pro Previewclinical-result-2211· Clinical Benchmarks data docs76.1%
146Claude Sonnet 5· Vals AI76.1%
147Claude Sonnet 5clinical-result-2212· Clinical Benchmarks data docs76.1%
148Gemini 2.5 Flash Lite Preview 09-2025· Vals AI75.8%
149Gemini 2.5 Flash-Liteclinical-result-2213· Clinical Benchmarks data docs75.8%
150Ling 3.0 Flash Finclinical-result-2214· Clinical Benchmarks data docs75.6%

Showing 101–150 of 207 results.

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