# MMBench-GUI

Four-level GUI-agent benchmark across six platforms, from content understanding and element grounding to task automation and collaboration, scored with an efficiency-quality area metric.

- Catalog URL: https://awesomecua.github.io/resources/github-open-compass-mmbench-gui/
- Original source: https://github.com/open-compass/MMBench-GUI
- Kind: benchmark
- Review: curated
- Scope: core
- Environments: cross-platform
- Topics: evaluation, grounding
- Evidence checked: 2026-10-16T06:35:58Z
- ID: github:open-compass/mmbench-gui

## Limitations

Inclusion does not establish reproducibility or endorsement.

- Record JSON: https://awesomecua.github.io/resources/github-open-compass-mmbench-gui/record.json

## Structured profile

- actions: not recorded
- evaluation: Level-specific evaluation; Efficiency–Quality Area (EQA) combines navigation quality and efficiency.
- focus: screen understanding, grounding, task automation, task collaboration
- observations: screenshots
- setting: mixed

### Count observations

- > 8,000 tasks — reported multi-level benchmark (reported; source: https://raw.githubusercontent.com/open-compass/MMBench-GUI/HEAD/README.md)

The README documents four levels, but its release notes and TODOs distinguish released L1/L2 evaluation from pending L3/L4 integration. Do not treat every level as the same evaluator or a complete implemented interface.

## License declarations

- code: not recorded
- data: not recorded
- document: not recorded
- weights: not recorded

## Primary evidence

- https://raw.githubusercontent.com/open-compass/MMBench-GUI/HEAD/README.md (retrieved 2026-10-16T06:35:58Z; SHA-256 35db9fca3ed1c1a3513924fefff4ccd224bc70cd7eb4bf8c54fcb9048851988e)