liteparse

Python, LlamaIndex, LangChain, Docker Compose: 15 Property Graph, 4 RDF , 10 Vector, OpenSearch, Elasticsearch, Alfresco, Nuxeo DBs. 14 data sources (10 auto-sync), KG auto-building, Ontologies, LLMs, Docling, LlamaParse, LiteParse, GraphRAG, RAG, Hybrid Search, AI Chat. TypeScript React, Vue, Angular frontends, REST, MCP Server, optional Langflow.

Do you need OCR or are heuristics enough? Find out!

Unified LiteParse fork and CPU-first research PDF-to-Markdown parser

Parameterised Docker images for LiteParse document parser — base, OCR, full, and API flavours with multi-arch support

Drop in scanned invoices — OCR'd 100% locally with LiteParse, audited by an LLM for math errors and duplicate charges, with every finding pinned to the exact spot on the scan.

Typed agentic RAG: every answer is a validated Pydantic object with verified verbatim quotes, chunk ids and a confidence score - or a deterministic refusal. LlamaIndex FunctionAgent + Nebius Token Factory + LiteParse.

MCP server for local PDF parsing with bounding boxes and visual citations, powered by LiteParse

litparse-rapidocr

DocMark Bench is an end-to-end Streamlit workbench for generating Markdown with several open-source document parsers, viewing the original beside every result, and ranking each conversion route for the current document.

ChunkLab is an open-source Streamlit app for learning, comparing, and benchmarking Retrieval-Augmented Generation (RAG) engineering techniques — chunking, embeddings, retrieval, reranking, and evaluation — side by side, with real metrics instead of vibes.

Streamlit application for translation using Cohere Labs Tiny Aya Global on Apple Silicon with MLX.

Extract text from any document locally. PaddleOCR + LiteParse v2. PDF, DOCX, XLSX, PPTX, images. Rust server, Docker Compose. Zero cloud, zero cost.