MedBench

What is MedBench?

MedBench v5 evaluates clinical multimodal models across 63 tasks with process-oriented metrics including stressors and hallucination tracking.

Released

2026-06-23

Evaluates

Agents, Biology & Drug Discovery, Multimodal

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.24155

MedBench paper

• Paper https://arxiv.org/abs/2606.24155

Which sources cite MedBench?

• MedBench

Related Agents, Biology & Drug Discovery, Multimodal benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• A Large-Scale Dataset and Benchmark
• A-Bench
• A-OKVQA
• A3S-Bench
• ABC-Bench
• ActionAtlas