Skip to content

Benchmark

MLCR-AA (Medical Long Context Reasoning)

Scores are percentages. Higher is better.

Models measured
25
Scores recorded
25
Sources
1
IQ weight
0.84

Score distribution

23.8%
Mean
22.2%
Median
10.3%
Std dev
10.6%
10th percentile
38.3%
90th percentile
6.1%
Min
6.1%41.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.84 for chat models, measured over 25 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 Sonnet 5.5clinical-result-2616· Clinical Benchmarks data docs41.7%
2GPT-6 Astraclinical-result-2614· Clinical Benchmarks data docs41.7%
3GPT-6.1 Solclinical-result-2615· Clinical Benchmarks data docs38.3%
4Kimi K3.1 Proclinical-result-2620· Clinical Benchmarks data docs38.3%
5Claude Haiku 5.5clinical-result-2623· Clinical Benchmarks data docs35%
6GPT-6 Solclinical-result-2621· Clinical Benchmarks data docs33.9%
7GPT-5.5 Instantclinical-result-2639· Clinical Benchmarks data docs31.7%
8GPT-6.1 Terraclinical-result-2617· Clinical Benchmarks data docs30.6%
9Claude Opus 5.5clinical-result-2613· Clinical Benchmarks data docs27.8%
10DeepSeek V4.1 Flash 1017clinical-result-2626· Clinical Benchmarks data docs26.7%
11GPT-6 Lunaclinical-result-2632· Clinical Benchmarks data docs26.1%
12Qwen 3.8 27Bclinical-result-2634· Clinical Benchmarks data docs22.8%
13Muse Wonderclinical-result-2622· Clinical Benchmarks data docs22.2%
14Muse Wonder Fullclinical-result-2618· Clinical Benchmarks data docs22.2%
15Grok 4.7clinical-result-2624· Clinical Benchmarks data docs20%
16Qwen3.8 Maxclinical-result-2625· Clinical Benchmarks data docs20%
17GLM 5.3clinical-result-2627· Clinical Benchmarks data docs17.2%
18MiniMax M3.1 Flashclinical-result-2631· Clinical Benchmarks data docs16.7%
19Claude Fable 5.1clinical-result-2619· Clinical Benchmarks data docs15.6%
20Ling 3.1 Flashclinical-result-2630· Clinical Benchmarks data docs15%
21Nemotron 3.5 Super VLclinical-result-2637· Clinical Benchmarks data docs13.3%
22MiMo V2.6 Proclinical-result-2629· Clinical Benchmarks data docs12.8%
23Gemini 3.8 Flashclinical-result-2628· Clinical Benchmarks data docs10.6%
24Muse Spark Openclinical-result-2633· Clinical Benchmarks data docs7.8%
25Inkling FP4clinical-result-2638· Clinical Benchmarks data docs6.1%

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