With 2,600+ GitHub stars since its May 2026 open-source launch, Knowhere solves the last-mile problem of document intelligence for AI systems — transforming complex unstructured PDFs, reports, and multi-page documents into structured JSON chunks that LLMs can consume without hallucination. The platform processes documents through an AI-native parsing pipeline that handles 20+ page documents with deep hierarchies, intricate tables, and multimodal content including images with OCR, achieving 95% precision in information extraction while reducing token costs by 50% compared to raw document ingestion. The knowledge tree architecture maintains historical context across multiple documents, enabling cross-document graph navigation for agentic retrieval that goes beyond simple chunk-based RAG. Built on Python 3.11+ with MinerU as the default PDF parser, the backend API runs alongside async workers that process document ingestion, graph construction, and embedding generation. The self-hosted Docker Compose stack packages the API server, processing workers, and Next.js dashboard for managing API keys, webhooks, and document-processing jobs, backed by PostgreSQL and Redis. Both Python and Node.js SDKs provide programmatic access for integration into existing AI pipelines and agent frameworks. LLM providers include DeepSeek and Alibaba Cloud DashScope with configurable key rotation for rate-limit management. Deploy on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

See exactly what your documents look like before a retrieval pipeline swallows them. Chunky puts the original PDF next to the converted Markdown with page-aware scrolling, so broken tables, scrambled columns, and stray artifacts are visible at a glance. Pick a converter, edit the Markdown by hand, then split it into color-coded chunks you can edit, delete, or enrich one at a time. Compare two converters on the same pages in a side-by-side dialog to find which one handles your layouts best. Try token, recursive, sentence, semantic, table, code, and heading-based splitters with adjustable size and overlap, and save the chunk sets that work. Optional model-powered enrichment adds titles, one-sentence context, summaries, keywords, and retrieval questions to each chunk. Process many documents in one batch and export everything as a ZIP. Judge chunk quality by eye, not by guesswork, before any of it reaches an embedding model. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.