Skip to content

Benchmark

Vals MedCode

Scores are percentages. Higher is better.

Models measured
99
Scores recorded
197
Sources
2
IQ weight
1.68

Score distribution

42%
Mean
42.4%
Median
9%
Std dev
31.3%
10th percentile
53.2%
90th percentile
19.7%
Min
19.7%63.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.68 for chat models, measured over 99 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GPT 5.5clinical-result-1978· Clinical Benchmarks data docs49.1%
52Kimi K3clinical-result-1979· Clinical Benchmarks data docs48.9%
53GPT-6.1 Solclinical-result-1980· Clinical Benchmarks data docs48.8%
54GPT-6.1 Sol· Vals AI48.8%
55GPT-6 Astraclinical-result-568· Clinical Benchmarks data docs48.5%
56GPT-6 Astra· Vals AI48.5%
57Claude Opus 4.6clinical-result-1981· Clinical Benchmarks data docs48.2%
58Gemini 3.8 Flash· Vals AI48.1%
59Gemini 3.8 Flashclinical-result-1982· Clinical Benchmarks data docs48.1%
60Gemini 3.1 Flash Lite· Vals AI47.6%
61Gemini 3.1 Flash Liteclinical-result-1983· Clinical Benchmarks data docs47.6%
62Claude Sonnet 5· Vals AI47.5%
63Claude Sonnet 5clinical-result-1984· Clinical Benchmarks data docs47.5%
64OpenAI o3· Vals AI47.3%
65OpenAI o3clinical-result-1985· Clinical Benchmarks data docs47.3%
66Claude Opus 4.1· Vals AI47.2%
67Claude Opus 4.1clinical-result-1986· Clinical Benchmarks data docs47.2%
68GPT-6 Sol· Vals AI47.1%
69GPT-6 Solclinical-result-1987· Clinical Benchmarks data docs47.1%
70MiniMax M3clinical-result-1988· Clinical Benchmarks data docs46.3%
71MiniMax M3· Vals AI46.3%
72Claude Opus 4.5clinical-result-1989· Clinical Benchmarks data docs45.2%
73MiMo V2.6 Proclinical-result-1990· Clinical Benchmarks data docs45%
74MiMo V2.6 Pro· Vals AI45%
75Grok 4.6· Vals AI44.7%
76Grok 4.6clinical-result-1991· Clinical Benchmarks data docs44.7%
77GPT-6 Lunaclinical-result-1992· Clinical Benchmarks data docs44.7%
78GPT-6 Luna· Vals AI44.7%
79Claude Sonnet 4.5· Vals AI44.1%
80Claude Sonnet 4.5clinical-result-1993· Clinical Benchmarks data docs44.1%
81Inkling 2 Small· Vals AI44.1%
82GPT-5.6 Sol· Vals AI44%
83GPT-5.6 Solclinical-result-1994· Clinical Benchmarks data docs44%
84Gemini 3.5 Flash Liteclinical-result-1995· Clinical Benchmarks data docs43.5%
85Gemini 3.5 Flash Lite· Vals AI43.5%
86GPT-5.6 Terra· Vals AI43.4%
87GPT-5.6 Terraclinical-result-1996· Clinical Benchmarks data docs43.4%
88Grok 4.5· Vals AI43.3%
89Grok 4.5clinical-result-1997· Clinical Benchmarks data docs43.3%
90Hy4 previewclinical-result-1998· Clinical Benchmarks data docs43.3%
91Hy4 preview· Vals AI43.2%
92GPT-5 miniclinical-result-1999· Clinical Benchmarks data docs43.1%
93GPT-5 mini· Vals AI43%
94GLM 5.3· Vals AI42.9%
95GLM 5.3clinical-result-2000· Clinical Benchmarks data docs42.9%
96DeepSeek V4 Proclinical-result-2001· Clinical Benchmarks data docs42.5%
97DeepSeek V4 Pro 0813· Vals AI42.5%
98GPT-5.6 Luna· Vals AI42.4%
99GPT-5.6 Lunaclinical-result-2002· Clinical Benchmarks data docs42.4%
100GLM 5.1· Vals AI41.6%

Showing 51–100 of 197 results.

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