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
1Claude Opus 5.5· Vals AI91.4%
2Claude Opus 5.5clinical-result-2151· Clinical Benchmarks data docs91.4%
3Claude Fable 5.1· Vals AI91.3%
4Claude Fable 5.1clinical-result-354· Clinical Benchmarks data docs91.3%
5Claude Sonnet 5.5· Vals AI91.1%
6Claude Sonnet 5.5clinical-result-2152· Clinical Benchmarks data docs91.1%
7Claude Opus 5· Vals AI91%
8Claude Opus 5clinical-result-150· Clinical Benchmarks data docs91%
9Inkling 2 Small· Vals AI90.2%
10Muse Spark 1.2· Vals AI90.1%
11Muse Spark 1.2clinical-result-355· Clinical Benchmarks data docs90.1%
12Claude Haiku 5.5· Vals AI89.5%
13Grok 4.7clinical-result-2153· Clinical Benchmarks data docs89.4%
14Grok 4.7· Vals AI89.4%
15Muse Wonder Full· Vals AI89.3%
16GLM 5.3 Flashclinical-result-2154· Clinical Benchmarks data docs88.9%
17GLM 5.3 Flash· Vals AI88.9%
18Muse Spark 1.1clinical-result-356· Clinical Benchmarks data docs88.9%
19Muse Spark 1.1· Vals AI88.9%
20GLM 5.3· Vals AI88.8%
21GLM 5.3clinical-result-2155· Clinical Benchmarks data docs88.8%
22Claude Fable 5· Vals AI88.5%
23Claude Fable 5clinical-result-357· Clinical Benchmarks data docs88.5%
24MiMo V2.6 Proclinical-result-2156· Clinical Benchmarks data docs88.3%
25MiMo V2.6 Pro· Vals AI88.3%
26GPT 5.1· Vals AI88.1%
27GPT 5.1clinical-result-358· Clinical Benchmarks data docs88.1%
28Kimi K3· Vals AI88%
29Kimi K3clinical-result-359· Clinical Benchmarks data docs88%
30GPT-6 Astraclinical-result-567· Clinical Benchmarks data docs87.9%
31GPT-6 Astra· Vals AI87.9%
32Gemini 4 Argonclinical-result-2393· Clinical Benchmarks data docs87.4%
33Gemini 4 Argon· Vals AI87.4%
34MiniMax M3· Vals AI87.3%
35MiniMax M3clinical-result-360· Clinical Benchmarks data docs87.3%
36Grok 4.5· Vals AI86.9%
37Grok 4.5clinical-result-2157· Clinical Benchmarks data docs86.9%
38GPT 5.5clinical-result-361· Clinical Benchmarks data docs86.9%
39GPT 5.5· Vals AI86.9%
40Claude Opus 4.6clinical-result-362· Clinical Benchmarks data docs86.7%
41Claude Opus 4.6· Vals AI86.7%
42Grok 4.6· Vals AI86.5%
43Grok 4.6clinical-result-363· Clinical Benchmarks data docs86.5%
44GPT-6.1 Sol· Vals AI86.5%
45GPT-6.1 Solclinical-result-2158· Clinical Benchmarks data docs86.5%
46Claude Opus 4.6clinical-result-2159· Clinical Benchmarks data docs86.1%
47Muse Spark· Vals AI85.9%
48Muse Sparkclinical-result-2160· Clinical Benchmarks data docs85.9%
49Claude Opus 4.8· Vals AI85.8%
50Claude Opus 4.8clinical-result-2161· Clinical Benchmarks data docs85.8%

Showing 1–50 of 207 results.

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