Skip to content

Benchmark

RCM-Bench-1

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
20
Scores recorded
20
Sources
1
IQ weight
1.89

Score distribution

86.9%
Mean
88.4%
Median
5.9%
Std dev
81.5%
10th percentile
91.2%
90th percentile
70.2%
Min
70.2%92.7%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.89 for chat models, measured over 20 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 5clinical-result-2566· Clinical Benchmarks data docs92.7%
2Gemini 4 Argonclinical-result-2567· Clinical Benchmarks data docs91.5%
3Claude Sonnet 5clinical-result-2568· Clinical Benchmarks data docs91.2%
4GPT-5.6 Solclinical-result-2569· Clinical Benchmarks data docs91.2%
5Claude Sonnet 5.5clinical-result-2570· Clinical Benchmarks data docs91.1%
6Gemini 3.8 Flashclinical-result-2571· Clinical Benchmarks data docs90.2%
7Claude Opus 5.5clinical-result-2572· Clinical Benchmarks data docs90.1%
8Claude Sonnet 4.6clinical-result-2573· Clinical Benchmarks data docs89.8%
9DeepSeek V4.1 Flashclinical-result-2574· Clinical Benchmarks data docs89%
10Claude Haiku 5.5clinical-result-2575· Clinical Benchmarks data docs88.4%
11GLM 5.3clinical-result-2576· Clinical Benchmarks data docs88.3%
12Grok 4.7clinical-result-2577· Clinical Benchmarks data docs88.2%
13Claude Opus 4.6clinical-result-2578· Clinical Benchmarks data docs87.8%
14GPT-6.1 Solclinical-result-2579· Clinical Benchmarks data docs87.4%
15GPT-5.6 Terraclinical-result-2580· Clinical Benchmarks data docs87%
16GPT-6 Lunaclinical-result-2581· Clinical Benchmarks data docs85.8%
17GPT-5 miniclinical-result-2582· Clinical Benchmarks data docs84.3%
18GPT-6 Solclinical-result-2583· Clinical Benchmarks data docs81.5%
19GPT-5.6 Lunaclinical-result-2584· Clinical Benchmarks data docs71.5%
20Claude Haiku 4.5clinical-result-2585· Clinical Benchmarks data docs70.2%

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