haiku.rag

Agentic RAG that answers questions about your own documents with citations to page numbers and section headings. Runs locally on an embedded database, no server required.

Built on LanceDB, Pydantic AI, and Docling. Full documentation at ggozad.github.io/haiku.rag.

Features

• Hybrid search — Vector + full-text with Reciprocal Rank Fusion
• Multimodal & cross-modal search — Multimodal embedders (vLLM, VoyageAI, Cohere) put picture vectors in the same space as text; supports text-as-query → figure hits and image-as-query
• Question answering — RAG capability with citations (page numbers, section headings)
• Vision QA — Vision-capable models receive figure bytes alongside chunk text; attach your own images to questions in ask and the chat TUI
• Reranking — local cross-encoders, Cohere, Zero Entropy, or vLLM
• Code execution — The RAG capability runs sandboxed Python over the documents for aggregation, computation and multi-document analysis
• Evidence compaction — Optional capability that replaces earlier questions' search results on the request with the evidence they cited, so long conversations stop resending everything they retrieved
• Citation policy — Optional capability that requires every answer to declare what grounds it, including declaring that nothing does
• Conversational RAG — Chat TUI and web application for multi-turn conversations with session memory
• Document structure — Stores full DoclingDocument, enabling structure-aware context expansion
• Multiple providers — Embeddings: Ollama, OpenAI, VoyageAI, Cohere, LM Studio, vLLM (multimodal via multimodal: true on vLLM/VoyageAI/Cohere). QA: any model supported by Pydantic AI
• Multi-database search — Search, ask, or chat across named databases with source attribution on results and citations
• Local-first — Embedded LanceDB, no servers required. Also supports S3, GCS, Azure, and LanceDB Cloud
• CLI & Python API — Full functionality from command line or code
• MCP server — Expose as tools for AI assistants (Claude Desktop, etc.)
• Visual grounding — View chunks highlighted on original page images
• Production ingester — Long-lived haiku-ingester service with persistent SQLite queue, async worker pool with retries and a dead-letter queue, FS / HTTP / S3 / WebDAV source adapters, FastAPI control plane, and a browser dashboard for operators. See docs/ingester.md.
• Tags — Name database states with haiku-rag tag and roll back to them
• Inspector — TUI for browsing documents, chunks, and search results

Installation

Python 3.12 or newer required

Full Package (Recommended)

pip install haiku.rag

Includes all features: document processing, all embedding providers, and rerankers.

Using uv? uv pip install haiku.rag

Slim Package (Minimal Dependencies)

pip install haiku.rag-slim

Install only the extras you need. See the Installation documentation for available options.

Quick Start

Note: Requires an embedding provider (Ollama, OpenAI, etc.). See the Tutorial for setup instructions.

# Index a PDF
haiku-rag add-src paper.pdf

# Search
haiku-rag search "attention mechanism"

# Ask questions with citations
haiku-rag ask "What datasets were used for evaluation?"

# Ask about an image (vision-capable model)
haiku-rag ask "Does this figure match the spec in the design doc?" --image figure.png

# Analyze — complex analytical tasks via code execution
haiku-rag ask "How many documents mention transformers?"

# Interactive chat — multi-turn conversations with memory
haiku-rag chat

# Continuously ingest from configured sources (FS, HTTP, S3, WebDAV)
haiku-ingester serve

See Configuration for customization options.

Python API

from haiku.rag.client import HaikuRAG

async with HaikuRAG("knowledge.lancedb", create=True) as rag:
    # Index documents
    await rag.create_document_from_source("paper.pdf")
    await rag.create_document_from_source("https://arxiv.org/pdf/1706.03762")

    # Search — returns chunks with provenance
    results = await rag.search("self-attention")
    for result in results:
        print(f"{result.score:.2f} | p.{result.page_numbers} | {result.content[:100]}")

    # QA with citations
    answer, citations = await rag.ask("What is the complexity of self-attention?")
    print(answer)
    for cite in citations:
        print(f"  [{cite.chunk_id}] p.{cite.page_numbers}: {cite.content[:80]}")

For direct agent composition, see the capabilities documentation.

MCP Server

Use with AI assistants like Claude Code, Codex, and Claude Desktop:

haiku-rag mcp --stdio

In Claude Code, install the plugin, which registers the server and a skill:

claude plugin marketplace add ggozad/haiku.rag
claude plugin install haiku-rag

In Codex, install the same plugin from its marketplace:

codex plugin marketplace add ggozad/haiku.rag
codex plugin add haiku-rag@haiku-rag

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "haiku-rag": {
      "command": "haiku-rag",
      "args": ["mcp", "--stdio"]
    }
  }
}

Provides search, document reading, and analysis tools directly in your AI assistant.

Examples

See the examples directory for working examples:

• Docker Setup - Complete Docker deployment with continuous ingestion (haiku-ingester) and MCP server
• Web Application - Full-stack conversational RAG with CopilotKit frontend

Documentation

Full documentation at: https://ggozad.github.io/haiku.rag/

• Quickstart - Provider setup and first ingestion
• Installation - Packages and extras
• Configuration - YAML reference
• CLI - Command reference
• Python API - Complete API docs
• Capabilities - Native Pydantic AI capabilities
• Tuning - Retrieval and answer-quality tuning
• Ingester - Production ingester for continuous indexing from FS, HTTP, S3, and WebDAV
• MCP - Model Context Protocol integration
• Remote processing - Offload conversion to docling-serve
• Applications - Chat TUI, web app, and inspector
• Benchmarks - Performance benchmarks
• Changelog - Version history

License

This project is licensed under the MIT License.

mcp-name: io.github.ggozad/haiku-rag