Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
Python ai4sciencedocument-analysisdocxextract-datalayout-analysisocrparserpdfpdf-converterpdf-extractor-llmpdf-extractor-pretrainpdf-extractor-ragpdf-parserpptxpythonxlsx
Commits on master
test(release): isolate template fixtures from published version
5dd3b38 myhloli committed 22 days ago
test(runtime): use complete llama selection in shutdown regression
58c0d64 myhloli committed 22 days ago
fix(tiers): update CPU core requirements for tier recommendations
b832b82 myhloli committed 22 days ago
chore: sync latest stable release in issue template
70d2841 github-actions[bot] committed 22 days ago
build: update docvortex and mineru-llama-cpp version constraints
abdc1fd myhloli committed 22 days ago
feat: update engine options to include llama-cpp and adjust documentation
bab3dc9 myhloli committed 22 days ago
refactor(ocr): unify degenerate crop handling and reduce copies
99e66df myhloli committed 22 days ago
fix(runtime): match Windows GPU memory by physical PCI identity
2502129 myhloli committed 22 days ago
feat(threading): implement default thread policy for GPU models
3c60e6f myhloli committed 22 days ago
refactor(runtime): reuse small-model device decisions for llama.cpp
169cf60 Xiaomeng Zhao committed 22 days ago
fix(pdf): replace residual private-use text with spaces after OCR
2c435eb myhloli committed 22 days ago
feat(gradio): unify automatic startup and default tier selection
955be0a myhloli committed 22 days ago