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
151DeepSeek V4 Pro· Vals AI75.1%
152DeepSeek V4 Proclinical-result-2215· Clinical Benchmarks data docs75.1%
153Grok 4.3clinical-result-2216· Clinical Benchmarks data docs74.4%
154Grok 4.3· Vals AI74.4%
155Claude Opus 4.1· Vals AI73.9%
156Claude Opus 4.1clinical-result-2217· Clinical Benchmarks data docs73.9%
157Gemini 2.5 Pro· Vals AI73.6%
158Gemini 2.5 Proclinical-result-2218· Clinical Benchmarks data docs73.5%
159GPT-5 nano· Vals AI72.9%
160GPT-5 nanoclinical-result-2219· Clinical Benchmarks data docs72.9%
161Gemini 2.5 Flash-Lite· Vals AI72.8%
162Gemini 2.5 Flash-Liteclinical-result-2220· Clinical Benchmarks data docs72.8%
163Qwen3 Maxclinical-result-2221· Clinical Benchmarks data docs72.7%
164Qwen3 Max· Vals AI72.7%
165Claude Sonnet 4· Vals AI72.4%
166Claude Sonnet 4clinical-result-2222· Clinical Benchmarks data docs72.4%
167GLM 5.1· Vals AI72.3%
168GLM 5.1clinical-result-2223· Clinical Benchmarks data docs72.3%
169MiMo V2.5· Vals AI72.2%
170MiMo V2.5clinical-result-2224· Clinical Benchmarks data docs72.2%
171Gemini 3 Pro Previewclinical-result-2225· Clinical Benchmarks data docs72%
172Gemini 3 Pro Preview· Vals AI72%
173Claude Opus 4.1clinical-result-2226· Clinical Benchmarks data docs71.8%
174Gemini 3.5 Flash Liteclinical-result-2227· Clinical Benchmarks data docs70.9%
175Gemini 3.5 Flash Lite· Vals AI70.9%
176Qwen3.5 Flashclinical-result-2228· Clinical Benchmarks data docs70.6%
177Qwen3.5 Flash· Vals AI70.6%
178Gemini 3 Flashclinical-result-2229· Clinical Benchmarks data docs69.9%
179Gemini 3 Flash· Vals AI69.9%
180Claude Sonnet 4thinking· Vals AI69.4%
181Claude Sonnet 4clinical-result-2230· Clinical Benchmarks data docs69.3%
182o4 Miniclinical-result-2231· Clinical Benchmarks data docs69.1%
183o4 Mini· Vals AI69.1%
184GLM 4.7clinical-result-2232· Clinical Benchmarks data docs68.6%
185GLM 4.7· Vals AI68.6%
186Mistral Medium 3.5clinical-result-2233· Clinical Benchmarks data docs67.7%
187Mistral Medium 3.5· Vals AI67.7%
188Gemini 2.5 Flash-Liteclinical-result-2234· Clinical Benchmarks data docs66.9%
189Gemini 2.5 Flash Lite Preview 09-2025thinking· Vals AI66.9%
190Laguna M.1· Vals AI65.9%
191Laguna M.1clinical-result-2235· Clinical Benchmarks data docs65.9%
192Gemini 3.1 Flash Lite· Vals AI63.9%
193Gemini 3.1 Flash Liteclinical-result-2236· Clinical Benchmarks data docs63.9%
194Grok 4.20· Vals AI63.4%
195Grok 4.20clinical-result-2237· Clinical Benchmarks data docs63.4%
196Laguna XS.2· Vals AI61.4%
197Laguna XS.2clinical-result-2238· Clinical Benchmarks data docs61.4%
198Command A Plus 05 2026· Vals AI55.7%
199Command A Plusclinical-result-2239· Clinical Benchmarks data docs55.7%
200Mercury 2.5· Vals AI55.1%

Showing 151–200 of 207 results.

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