Vibe Coding Discovermodel-compose
View on GitHubDeploy production-ready AI services in minutes. One YAML file for agents, RAG pipelines, and MCP servers — run anywhere. Inspired by docker-compose.
model-compose is a declarative YAML orchestrator (docker-compose for AI) that deploys chat APIs, ReAct agents, RAG pipelines, and MCP servers from one file. It bridges local and cloud models and runs on Docker, native, or distributed Redis-queued runtimes.
Use Cases
Deploy a chat API from one YAML fileBuild declarative RAG pipelinesConfigure ReAct agents with tools in YAMLExpose workflows as MCP servers for Claude/CursorStream tokens, audio, and video frames end-to-endRun local models alongside cloud APIsScale AI services horizontally via Redis queuesHuman-in-the-loop workflow approvalReal-time TTS and media processing pipelinesWeb scraping and browser automation workflows
Built With
- Language
- Python
- Frameworks
- FastAPI · Uvicorn · Pydantic · Click · aiohttp · Gradio · MCP SDK · Docker · Chroma · Milvus · Qdrant · FAISS · Neo4j · Redis
Tags
yaml · declarative · orchestration · llmops · agents · rag · mcp · multi-provider · docker · streaming · workflow · self-hosted · human-in-the-loop · vector-database · local-models · docker-compose-alternative