A chat agent web UI that supports Google search, file and URL conversation (RAG), and code interpreter functionality, recreating features similar to Kimi Chat.
One-Minute Overview#
ChatPilot is a comprehensive Chat Agent web interface that supports multiple large language model integrations, providing features like Google search, file and URL conversation (RAG), and code interpreter functionality. It's suitable for developers, researchers, and enterprises building their own conversational AI assistants, especially for scenarios requiring document analysis, code execution, and web search capabilities.
Core Value: A one-stop AI conversation solution integrating search, document processing, and code execution capabilities
Getting Started#
Installation Difficulty: Medium - Requires API key configuration and environment variables, but Docker deployment is simplified
# Quick start with Docker
export OPENAI_API_KEY=sk-xxx
export OPENAI_BASE_URL=https://xxx/v1
docker run -it -e OPENAI_API_KEY=$OPENAI_API_KEY -e OPENAI_BASE_URL=$OPENAI_BASE_URL -p 8080:8080 shibing624/chatpilot:0.0.1
Is this suitable for me?
- ✅ Need an AI assistant with document analysis and search integration: Supports RAG functionality for document Q&A
- ✅ Need code interpretation and execution environment: Built-in Python code interpreter
- ❌ Simple chatbot applications: May be overly complex for basic use cases
Core Capabilities#
1. Tool Calling System - Expanding AI Capabilities#
- Supports web search tools (Google Search API), URL auto-parsing tools, and Python code interpreter Actual Value: Breaks through AI knowledge limitations with real-time information retrieval and code execution
2. Multi-Model Integration - Flexible Model Selection#
- Supports OpenAI/Azure API for GPT series models, Ollama API for local open-source models, and litellm API for various cloud service deployments Actual Value: Choose the optimal cost-effective model based on needs, supporting both private deployment and hybrid cloud architectures
3. RAG File Q&A - Enhanced Knowledge Processing#
- Supports Agent calls for RAG file Q&A, handling long documents and professional knowledge Q&A Actual Value: Resolves large language model context limitations and improves accuracy in professional domain Q&A
4. Frontend-Backend Separation - Easy Customization and Extension#
- Frontend uses Svelte, backend uses FastAPI, supporting separated deployment Actual Value: Facilitates team development with independent frontend and backend service upgrades
5. Multimedia Support - Enriched Interaction Experience#
- Supports voice input/output and image generation Actual Value: Provides more natural interaction methods for diverse application scenarios