knowhere
Knowhere extracts, parses, and outputs structured chunks ready for AI Agents and RAG.
What is knowhere?
Knowhere extracts, parses, and outputs structured chunks ready for AI Agents and RAG.
Overview of knowhere
What it is
Knowhere is a document intelligence and retrieval infrastructure that turns complex, dirty files into persistent, navigable memory for AI agents. It ingests unstructured documents and produces persistent, navigable memory: parsing, hierarchy reconstruction, multi-modal structuring, and graph construction in a single pipeline. Every result stays connected to its document, section, source pages, and related assets. It parses files along their native hierarchy and uses frontier vision models to cut extraction errors. Documents can be ingested progressively: readable and searchable first, enriched in the background.
Key features
- Route by format: Page-based documents (PDF, PowerPoint) use vision parsing; everything else is parsed natively. Both end up in the same chunk and metadata schema.
- Progressive ingestion: With
ingestion_mode=progressive, a document becomes readable and searchable first, then gains its outline and hierarchy, and is finally enriched in the background. - Recall without brittle reconstruction: Pages can be indexed and searched through summaries, entities, source text, and hierarchy even when OCR or layout extraction cannot reliably recover every component.
- One navigable memory: Text sections and vision pages become compatible hierarchy nodes with source evidence, linked assets, and cross-document relationships.
- Agentic Retrieval: Knowhere provides corpus navigation tools (recall, grep, outline, read) that agents can use to explore the document memory. External agents can call these tools via MCP or API.
- Supports ultra-long PDFs and atlases (e.g., over 1000 pages) and routes technical atlases through a dedicated layout-aware parser.
- Supports multiple image formats: .gif, .webp, .bmp, .tif/.tiff, .heic/.heif.
Use cases
Knowhere is designed for AI agents and RAG (Retrieval-Augmented Generation) workflows. It can be used for agentic RAG, vector-based RAG, or any LLM workflow that requires structured, navigable document memory. It is suitable for processing complex, dirty files such as PDFs, slide decks, and technical atlases, and for building persistent memory that agents can query with hierarchy-aware tools.
Getting started
See the repository README for installation and quick-start instructions. Knowhere Cloud is available at knowhereto.ai with $5 in free credits upon registration. For self-hosting, refer to the knowhere-self-hosted repository.
License and pricing
Knowhere is open source under the Apache-2.0 license. Pricing is not stated beyond being open source; the managed cloud service offers free credits.
Based on the project's public README and repository metadata.