RiceBall is an open-source, full-stack AI Agent & Knowledge Base Platform. It aims to help teams and enterprises quickly build and deploy LLM-based intelligent applications in a private environment.
Core Capabilities:
- Private RAG Knowledge Base: Supports document upload, automatic chunking, and vectorization, enabling AI to answer questions based on your private data.
- Agent Engine: Based on LangChain, supports tool calling and MCP (Model Context Protocol) (🚧 In Progress), empowering AI to execute tasks.
- Multi-Model Aggregation: Supports mainstream interface protocols like OpenAI and Anthropic, avoiding vendor lock-in.
- Modern Full-Stack Architecture: Backend uses FastAPI (Python), frontend uses Nuxt 3 (Vue), with built-in OAuth authentication.
In the process of AI adoption, enterprises often face the dilemma of balancing data security and flexibility. RiceBall provides best practices:
- Complete Data Control: Supports local deployment (Docker); all data (knowledge base, chat history) is stored on your private server.
- Deep Business Integration: Through tool calling and the MCP protocol (🚧 In Progress), RiceBall can connect to your databases, APIs, and internal tools, becoming a true business assistant.
- Flexible Model Strategy: Choose models based on scenarios—use high-performance models for complex reasoning, and cost-effective models for daily conversation to optimize costs.
- Developer Friendly: Provides a clear modular architecture and comprehensive APIs, facilitating secondary development and customization.
- Enterprises & Teams: Building internal knowledge base assistants, intelligent customer service, and R&D efficiency tools.
- Full-Stack Developers: Developers looking for a mature RAG + Agent architecture as a starting point.
- System Integrators: Service providers delivering private AI solutions to clients.
The name RiceBall comes from a hamster I own. Every time it eats in its little food bowl, it curls up like a rice ball, so I named it RiceBall.
Note: This setup is for preview and testing purposes only. Please use with caution in production environments.
Run RiceBall with a single command using our All-in-One Docker image (includes SQLite & Local Storage):
docker run -d \
-p 8000:8000 \
-e SUPERUSER_EMAIL=admin@admin.com \
-e SUPERUSER_PASSWORD=admin123456 \
-v riceball_storage:/app/storage \
--name riceball \
ghcr.io/riceball-ai/riceball:all-in-one-latestVisit http://localhost:8000 to start using RiceBall. You can configure the initial superuser credentials via the SUPERUSER_EMAIL and SUPERUSER_PASSWORD environment variables.
For production environments requiring PostgreSQL and S3, you can deploy from source:
git clone https://github.com/riceball-ai/riceball.git
cd riceball
docker compose -f docker-compose.prod.yml up -d
RiceBall wouldn't exist without the contributions of the open-source community. Special thanks to the following excellent open-source projects:
It is impossible to list all projects. If your project is used but not listed here, please contact us to add it.
This project is open source under the MIT License.