Skip to content

Django RAGKit

Production-ready Retrieval-Augmented Generation (RAG) and AI Chat toolkit for Django with PostgreSQL & pgvector.

What is Django RAGKit?

Django RAGKit is a batteries-included Django application designed to easily integrate Retrieval-Augmented Generation (RAG) and intelligent conversational AI into any Django project.

Rather than stitching together separate vector databases, external orchestrators, and custom API glue code, Django RAGKit keeps your data and knowledge base directly in your primary PostgreSQL database using the native pgvector extension. It provides everything from vector storage and HNSW cosine indexing to modular LLM/Embedding providers, automatic signals for embedding synchronization, and a built-in interactive chat interface.

graph LR
    User([User]) -->|Ask Question| ChatUI[Chat Interface / API]
    ChatUI -->|Query| Pipeline[RAG Pipeline]
    Pipeline -->|Generate Query Vector| Embed[Embedding Provider]
    Pipeline -->|HNSW Cosine Search| PG[(PostgreSQL + pgvector)]
    PG -->|Relevant Q&A Context| Pipeline
    Pipeline -->|Synthesize Prompt| LLM[LLM Provider]
    LLM -->|Stream / Return Answer| Pipeline
    Pipeline -->|Store Question & Answer| PG
    Pipeline -->|Return Answer| ChatUI

Key Features

  • Native pgvector Integration: Stores embeddings directly in PostgreSQL using VectorField and accelerated by HnswIndex with cosine distance operations (vector_cosine_ops).
  • Modular Provider Architecture: Switch between local models (Ollama) and cloud endpoints (OpenRouter) without altering business logic. Easily extensible for custom providers.
  • Automated Embedding Synchronization: Django signals (post_save and pre_save) automatically compute embeddings when new Q&A items are added or updated in the admin panel or database.
  • Zero-Friction Embeddings Reset: Changing embedding models or dimensions? The python manage.py reset_embeddings command safely backs up existing tables, updates schema definitions, and executes migrations automatically.
  • Built-in Modern Web Chat UI: Ready-to-use interactive chat interface with session management via UUIDs, async message handling, and feedback logging.
  • Flexible Access Control: Toggle between guest-friendly chat and authenticated-only mode using built-in view mixins (RagkitLoginRequiredMixin, RagkitApiLoginRequiredMixin).

Quick Navigation

Section Description
Python Package (pip) Integrate into an existing Django project via pip
Docker Quickstart One-line installer and automated environment configuration
Docker Image Official pre-built Docker Hub image (shincuff/django-ragkit:latest)
Settings Reference Complete documentation of all RAGKIT configuration keys
RAG Pipeline Architecture Detailed walkthrough of the retrieval and generation lifecycle
Reset Embeddings Command Guide to schema migration when altering vector dimensions

License

Django RAGKit is open-source software licensed under the MIT License.