AI applications
MinerU-Diffusion
opendatalab/MinerU-Diffusion
A diffusion-based document OCR system that replaces autoregressive decoding with parallel block decoding to parse pages into structured markdown.
Topics
Related projects
mineru83.4kAn AI-powered document parsing app that converts PDFs and Office files into clean LLM-ready Markdown and JSON.
Dolphin9.1kDolphin is ByteDance's open document image parsing model, released with weights and inference code to convert PDFs and images into structured JSON and Markdown.
manga-image-translator10.6kAI-powered manga and image translator that detects text, erases it and renders translations, run locally or via Docker.
easydiffusion10.5kA one-click Stable Diffusion app that runs locally and serves a browser UI for generating, inpainting, and upscaling AI artwork from text prompts.
text-extract-api3.2ktext-extract-api is a FastAPI service that combines OCR engines and Ollama-hosted vision models to convert PDFs, Office files and images into Markdown or structured JSON, including PII anonymization.
retain-pdf2.7kAn AI-powered PDF translation tool that preserves layout and formulas in scanned or academic documents, offered as a desktop app and self-hosted Docker service.