Multimodal evaluation for endoscopy image and video analysis.

EndoBench is a benchmark in healthcare AI. Multimodal evaluation for endoscopy image and video analysis. Scores are reported as: Percent accuracy across endoscopy tasks.

Audited by MedCheckLiu et al., 2025

Key facts

What it measuresMultimodal evaluation for endoscopy image and video analysis.
Who should careGastroenterology and surgical AI teams.
Use it whenEndoscopy-specific model selection.
Do not use it forGeneral clinical evaluation.
Score scalePercent accuracy across endoscopy tasks.
Grader methodunknown
First released2025
LanguagesEnglish
Use casesimaging, multimodal
CitationLiu et al., 2025
Statusactive

Paper

Frequently asked questions

What does EndoBench measure?

Multimodal evaluation for endoscopy image and video analysis.

Who should use EndoBench?

Gastroenterology and surgical AI teams. Endoscopy-specific model selection.

What should EndoBench not be used for?

General clinical evaluation.

How are EndoBench scores reported?

Percent accuracy across endoscopy tasks. Scores from different benchmarks are not comparable to each other.

Is EndoBench independent?

No conflicts of interest have been verified for EndoBench in this registry as of 2026-08-05. Absence of a recorded conflict means none has been verified, not that none exists.

Compare with

Asclepius

Spectrum evaluation for multimodal medical models across many imaging modalities and difficulty levels.

CheXpert

A large chest radiograph dataset with uncertainty labels and expert comparison. Predates language models and i

GMAI-MMBench

Large multimodal benchmark for general medical AI, spanning many imaging modalities, departments and task type

MMMU (Health and Medicine)

The health and medicine subset of a broad expert-level multimodal reasoning benchmark.

OmniMedVQA

Very large medical visual question answering set spanning many modalities and anatomical regions.

PathMMU

Expert-level pathology understanding and reasoning from slide images.

Last verified 2026-08-05. Governance and conflict data compiled by Healthcare Discovery from primary sources, each linked above. Seed inventory from Ma et al., Beyond the Leaderboard, ACL 2026. How we verify · All 63 entities