unlimited-ocr

Convert PDF to markdown using a GUI/CLI for Unlimited-OCR. Uses GGUF models

Unlimited-OCR on Apple Silicon via MLX MXFP8 (mlx-vlm 0.6+). Includes config fix for native unlimited-ocr loading.

Reference Sliding Window Attention (R-SWA) from Baidu's Unlimited OCR, reproduced from scratch in PyTorch. Runs on Apple Silicon (MPS) and CPU — no CUDA. Constant KV cache, with benchmarks.

A high-performance document parsing and structuring pipeline using Baidu Unlimited-OCR and local Ollama.

Unlimited-OCR for DeepSeek Harness with a native tool and GUI configuration

Portable Agent Skill for Baidu Unlimited-OCR cloud and local long-document parsing

百度 Unlimited-OCR 模型的本地化服务封装：OpenAI 兼容 API + 单图/批量 OCR 端点

Free, open-source Google Colab notebook that converts PDF & image files (JPG, PNG, BMP, TIFF, WEBP) into clean Markdown using Baidu's Unlimited-OCR (MIT license, 93%+ OmniDocBench). No API key, no cost — runs on free Colab GPU. PDF to Markdown, OCR, document parsing, image to text converter.

Unlimited-OCR 的 OpenVINO 本地适配：显式 KV、R-SWA cache、12 层 sparse MoE decode、FP4/INT8 mixed experts

OCR Platform with RAG and hybrid query routing.

Windows desktop app for baidu/Unlimited-OCR that installs its own Python, PyTorch and 6.67 GB of weights — no terminal required. Runs fully offline on CPU or CUDA.