X-PLUG/mPLUG-DocOwl
A multimodal LLM family developed by Alibaba for understanding documents, charts, and tables without OCR.

Not currently ranked — collecting fresh signals.
star history
mPLUG-DocOwl is a series of multimodal large language models designed for document understanding tasks including chart comprehension, table extraction, and overall document analysis. The project includes multiple versions (DocOwl1.5, DocOwl2, TinyChart) with training and inference code, enabling users to fine-tune models on their own data for specialized document understanding applications.
Frequently asked
- What is X-PLUG/mPLUG-DocOwl?
- A multimodal LLM family developed by Alibaba for understanding documents, charts, and tables without OCR.
- Is mPLUG-DocOwl open source?
- Yes — X-PLUG/mPLUG-DocOwl is open source, released under the Apache-2.0 license.
- What language is mPLUG-DocOwl written in?
- X-PLUG/mPLUG-DocOwl is primarily written in Python.
- How popular is mPLUG-DocOwl?
- X-PLUG/mPLUG-DocOwl has 2.4k stars on GitHub.
- Where can I find mPLUG-DocOwl?
- X-PLUG/mPLUG-DocOwl is on GitHub at https://github.com/X-PLUG/mPLUG-DocOwl.