Converge the SDK, Gradio, remote CLI, Doclib and Router on Tasks with hash-based uploads, per-file result URLs, task listing and optional model output. Preserve Files and the synchronous file_parse wrapper while removing standalone Uploads/Jobs and local-path HTTP inputs.
Migrate contract, lifecycle, transport and real-document regression coverage; update documentation and HTTP examples. Validation: 1452 passed, 3 conditional skips, plus targeted final PDF checks.
BREAKING CHANGE: clients and workers must use Tasks; standalone Uploads/Jobs endpoints and local-source/inline HTTP options have been removed.
Accept form manifests and per-file uploads with resumable task state and optional source hashes. Publish result URLs only after final ZIP artifacts are ready, and align Router handling with the same protocol.
Remove synchronous parsing and task result routes, drop fixed local upload limits, and update the HTTP client, documentation, and regression coverage.
Add auto/cpu/gpu configuration, Windows adapter discovery, isolated rec calibration, atomic preference persistence, and vendor session presets. Route layout, rec, and full MFR batches through DirectML with CPU fallback.
Use mutually exclusive Windows x64 DirectML dependencies and guard ORT/Magika distribution identity. Include calibration fixtures, three-machine results, the 0.8 threshold with ten UHD 770 repeats, and installation documentation.
Validation: 762 MinerU tests and 71 DocVortex tests passed; dependency checks and Windows Python 3.10-3.14 resolution passed. The 19-page OCR result matches the CPU baseline. DocVortex PyPI acceptance remains pending a compatible release; pinned source acceptance passed.
- Deleted .readthedocs.yaml files from multiple directories
- Removed outdated API and user guide documentation files
- Deleted command line usage examples
- Removed CUDA acceleration guide
- Commented out the code that copies the paddleocr model to user directory
- This change affects both download_models.py and download_models_hf.py scripts
- Add PaddleOCR model to the list of models to download
- Update download_models.py and download_models_hf.py scripts to include PaddleOCR
- Create user directory for PaddleOCR and copy downloaded models
- Update config version to1.2.0
- Refactor model initialization in model_init.py- Update dependencies in requirements.txt files
- Remove unused imports and models
- Add conditional imports for table models
- Update model path from 'unimernet_small' to 'unimernet_small_2501' in multiple scripts and configuration files
- This change affects download_models.py, download_models_hf.py, and model_configs.yaml
- Update the version check in download_models.py and download_models_hf.py
- Change the threshold from '1.1.0' to '1.1.1' for model configuration updates
- Replace Gitee URLs with jsdelivr URLs in multiple files
- Update Dockerfile instructions to use new URLs
- Modify documentation to reflect new resource location
- Adjust scripts and READMEs to point to new URL for downloads
- Add LLM-aided formula and text correction functionality
- Update config reader to include LLM-aided settings
- Create new LLM-aided processing module
- Update main processing script to incorporate LLM-aided corrections
- Modify download scripts to check for new config version
- Remove the download script from the `docs` directory and unify it in the `scripts` directory.
- Update the JSON template URL in the script to the master branch.
- Modify the script download link provided in the documentation.
- Update the table recognition model in the configuration template to `rapid_table`.
myhloli