# WebJudge 7B

A visual trajectory judge that retains selected intermediate screenshots to evaluate whether a web agent completed its task.

- Catalog URL: https://awesomecua.github.io/resources/url-772161f06c2b3d4978b5/
- Original source: https://huggingface.co/osunlp/WebJudge-7B
- Kind: model
- Review: curated
- Scope: core
- Environments: web
- Topics: evaluation, reward-models, web-navigation
- Evidence checked: 2026-10-17T04:03:41Z
- ID: url:772161f06c2b3d4978b5

## Limitations

Four public safetensors shards are present. Apache-2.0 is declared for the weights. This is an evaluator checkpoint rather than a navigation policy; human-agreement rates are not treated as task-success scores.

- Record JSON: https://awesomecua.github.io/resources/url-772161f06c2b3d4978b5/record.json

## License declarations

- code: not recorded
- data: not recorded
- document: not recorded
- weights: apache-2.0 — https://huggingface.co/osunlp/WebJudge-7B/raw/main/README.md

## Primary evidence

- https://huggingface.co/osunlp/WebJudge-7B/raw/main/README.md (retrieved 2026-10-17T04:03:41Z; SHA-256 425f11651cdae09997c1b038cdc9c091778c35a51b8de88140fc6dc935b8bb64)
- https://huggingface.co/api/models/osunlp/WebJudge-7B (retrieved 2026-10-17T04:03:41Z; SHA-256 2eebc1f909263be946057306b41d733e29f0f32337dae02da4e989b5477cf31f)
- https://huggingface.co/osunlp/WebJudge-7B/raw/main/README.md (retrieved 2026-10-17T04:03:41Z; SHA-256 425f11651cdae09997c1b038cdc9c091778c35a51b8de88140fc6dc935b8bb64)