[Feature] 请问是否可以支持多模态模型donut的推理
Open
- Dominant language
- Python
- Stars
- 8.1k
- Forks
- 748
- Avg merge
- 6d 2h
- Merged PRs (30d)
- 54
Description
### Motivation
目前lmdeploy支持了很多VLMs，请问是否可以支持donut多模态模型，其基于vision-encoder-decoder。
### Related resources
donut：https://huggingface.co/docs/transformers/model_doc/donut
vision-encoder-decoder：https://huggingface.co/docs/transformers/model_doc/vision-encoder-decoder
Contributor guide
Assessment
This issue has not been assessed yet.