ChromaDB RAG Backend
by Kitworks Systems
850,00 €
| Odoo Apps Dependencies | base, base_automation, bus, http_routing, mail |
| Community Apps Dependencies | Show |
| Technical Name | kw_ai_rag_chroma |
| License | OPL-1 |
| Website | https://kitworks.systems/ |
| Versions | 17.0 18.0 19.0 |
AI Connector — RAG Backend
ChromaDB — Developer-Friendly Vector Backend
Open-source vector database designed for fast iteration. Run it embedded or as a standalone server — perfect for pilots and development environments.
HNSW
Index Algorithm
REST
API Transport
100%
Open Source
When to Use ChromaDB
Rapid prototyping
Spin up ChromaDB with a single Docker command. Build your first RAG pipeline in hours, not days.
Development and staging
Embedded mode stores vectors to disk with zero configuration. Move to client-server mode for staging without changing Odoo code.
Small to medium deployments
Up to ~500 k vectors with consistent performance. Straightforward migration path to Qdrant when scaling further.
Technical Specifications
| Property | Value | Notes |
|---|---|---|
| ChromaDB version | 0.4+ | chromadb pip package or Docker image |
| Index type | HNSW (hnswlib) | Automatic, no tuning required |
| Similarity metric | Cosine / L2 / IP | Set per collection |
| Deployment modes | Embedded / HTTP | Configurable in Odoo AI settings |
| Recommended scale | Up to 500 k vectors | Beyond this, migrate to Qdrant |
Requirements
- ✓ kw_ai — AI Connector core
- ✓ kw_ai_rag — RAG orchestration module
- ✓ chromadb Python package (embedded) or ChromaDB server (Docker)
- ✓ Any embedding module
Start your RAG pilot today
Developer-friendly setup. No infrastructure expertise required.
Book a Demo