Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Visual Question Answering on CMMMU

26.1Accuracy

DAVE

Updated 1mo ago

Evaluation Results

Method	Links
DAVE 2025.12		26.1
SigLIP 2 2025.12		21.4

SOTA Paper

DAVE

DAVE: A VLM Vision Encoder for Document Understanding and Web Agents

Dataset

CMMMU

Follow for update

@wizwand_team Discord

Related Benchmarks

Web UI Question Answering on VisualWebBench Web UI Localization on Screenspot V2 Web UI Question Answering on WebSRC Document Layout Analysis on DocBank UI Screen Classification on RICO SCA

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs