# PC-Agent-E checkpoint

Computer-use checkpoint fine-tuned on PC-Agent-E trajectories; the model-card prose identifies Qwen2.5-VL-72B-Instruct as its starting model.

- Catalog URL: https://awesomecua.github.io/resources/url-04aa7569d183379d615c/
- Original source: https://huggingface.co/henryhe0123/PC-Agent-E
- Kind: model
- Review: curated
- Scope: core
- Environments: desktop, windows, linux
- Topics: action-prediction, imitation-learning
- Evidence checked: 2026-10-18T04:35:19Z
- ID: url:04aa7569d183379d615c

## Limitations

Public manifest lists 31 Safetensors shards and their index. The card prose names Qwen2.5-VL-72B-Instruct while its YAML base_model field self-references PC-Agent-E; no base-model relationship is imported from that inconsistent metadata. The model-card MIT declaration is recorded for weights only.

- Record JSON: https://awesomecua.github.io/resources/url-04aa7569d183379d615c/record.json

## License declarations

- code: not recorded
- data: not recorded
- document: not recorded
- weights: MIT — https://huggingface.co/henryhe0123/PC-Agent-E/raw/568ab45861d5dab041e158d47211c6f53ba343f2/README.md

- weights review: Publisher model-card declaration for the checkpoint; code and training-data terms are separate.
  Sources: https://huggingface.co/henryhe0123/PC-Agent-E/raw/568ab45861d5dab041e158d47211c6f53ba343f2/README.md

## Primary evidence

- https://huggingface.co/henryhe0123/PC-Agent-E/raw/568ab45861d5dab041e158d47211c6f53ba343f2/README.md (retrieved 2026-10-18T04:35:19Z; SHA-256 feb41002efff5ef30fe06f9891efbe954e0d046c0379f7c8ca72b5d8a8a304be)
- https://huggingface.co/api/models/henryhe0123/PC-Agent-E (retrieved 2026-10-18T04:33:21Z; SHA-256 f81a6cea4234e65f8bab390a18487e65dd0b0cb62f0a9458c658ff07e8a7b5f7)
- https://raw.githubusercontent.com/GAIR-NLP/PC-Agent-E/HEAD/README.md (retrieved 2026-10-18T04:27:13Z; SHA-256 3ffb8c843c6acb8c2d5bbba11ba49d6b36cd566553a89974b6252a2bfd289d5d)