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
1Claude Opus 5· Vals AI63.6%
2Claude Opus 5clinical-result-145· Clinical Benchmarks data docs63.6%
3Gemini 3.1 Pro Preview· Vals AI59.1%
4Gemini 3.1 Pro Previewclinical-result-146· Clinical Benchmarks data docs59.1%
5Gemini 4 Argon· Vals AI58.8%
6Gemini 4 Argonclinical-result-2392· Clinical Benchmarks data docs58.8%
7Claude Fable 5· Vals AI56.1%
8Claude Fable 5clinical-result-147· Clinical Benchmarks data docs56.1%
9Gemini 3 Flash· Vals AI55.9%
10Gemini 3 Flashclinical-result-148· Clinical Benchmarks data docs55.9%
11Gemini 3.5 Flashclinical-result-149· Clinical Benchmarks data docs55.8%
12Gemini 3.5 Flash· Vals AI55.8%
13Claude Opus 4.7clinical-result-346· Clinical Benchmarks data docs54.9%
14Claude Opus 4.7· Vals AI54.9%
15Claude Fable 5.1clinical-result-348· Clinical Benchmarks data docs53.5%
16Claude Fable 5.1· Vals AI53.5%
17Gemini 3.7 Flash· Vals AI53.4%
18Gemini 3.7 Flashclinical-result-1966· Clinical Benchmarks data docs53.4%
19Claude Opus 4.8clinical-result-350· Clinical Benchmarks data docs53.2%
20Claude Opus 4.8· Vals AI53.2%
21Gemini 3.6 Flash· Vals AI53.2%
22Gemini 3.6 Flashclinical-result-352· Clinical Benchmarks data docs53.2%
23Claude Sonnet 5.5· Vals AI52.9%
24Claude Sonnet 5.5clinical-result-1967· Clinical Benchmarks data docs52.9%
25GPT 5.1· Vals AI52.7%
26GPT 5.1clinical-result-353· Clinical Benchmarks data docs52.7%
27Gemini 3 Pro Previewclinical-result-1968· Clinical Benchmarks data docs52.2%
28Gemini 3 Pro Preview· Vals AI52.2%
29Claude Haiku 5.5· Vals AI51.5%
30Muse Spark· Vals AI51.3%
31Muse Sparkclinical-result-1969· Clinical Benchmarks data docs51.3%
32Gemini 2.5 Pro· Vals AI50.6%
33Gemini 2.5 Proclinical-result-1970· Clinical Benchmarks data docs50.6%
34Muse Wonder Full· Vals AI49.9%
35Claude Opus 5.5clinical-result-1971· Clinical Benchmarks data docs49.8%
36Claude Opus 5.5· Vals AI49.8%
37GPT 5.2clinical-result-1972· Clinical Benchmarks data docs49.8%
38GPT 5.2· Vals AI49.7%
39GPT 5· Vals AI49.6%
40GPT 5clinical-result-1973· Clinical Benchmarks data docs49.6%
41Grok 4.7· Vals AI49.6%
42Grok 4.7clinical-result-1974· Clinical Benchmarks data docs49.6%
43Kimi K3· Vals AI49.4%
44Muse Spark 1.2clinical-result-1975· Clinical Benchmarks data docs49.4%
45Muse Spark 1.2· Vals AI49.3%
46Claude Opus 4.5clinical-result-1976· Clinical Benchmarks data docs49.2%
47Claude Opus 4.5· Vals AI49.2%
48Claude Opus 4.6clinical-result-1977· Clinical Benchmarks data docs49.1%
49Claude Opus 4.6· Vals AI49.1%
50GPT 5.5· Vals AI49.1%

Showing 1–50 of 197 results.

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