Benchmark

MedXpertQA

A comprehensive benchmark to evaluate expert-level medical knowledge and advanced reasoning, featuring 4,460 questions spanning 17 specialties and 11 body…

Modality
multimodal
Categories
multimodal, reasoning, healthcare, vision
Openness
unknown
Source
llm_stats
Reported scores
12

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
DiffusionGemma 26B-A4BGoogle0.492026-06-10evidence
Gemma 4 12BGoogle0.4872026-05-23evidence
Gemma 4 26B-A4BGoogle0.5812026-04-02evidence
Gemma 4 31BGoogle0.6132026-04-02evidence
Gemma 4 E2BGoogle0.2352026-04-02evidence
Gemma 4 E4BGoogle0.2872026-04-02evidence
MAI-Thinking-1Microsoft0.432026-06-02evidence
MedGemma 4B ITGoogle0.1882025-05-20evidence
Muse SparkMeta0.7842026-04-08evidence
Qwen3.5-122B-A10BQwen0.6732026-02-24evidence
Qwen3.5-27BQwen0.6242026-02-24evidence
Qwen3.5-35B-A3BQwen0.6142026-02-24evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.