YaoApp/yao

Project overview

yaoapp/yao consolidates what typically requires a complex multi-runtime stack into a single binary, removing the need to coordinate separate environments for building and deploying AI agents and full web applications together.

Project type: MCP · AI Agent · RAG
Use cases: Knowledge Q&A · Automation
Deployment: Refer to project documentation
License: License pending

Best for

Developers and AI engineers who want to build and run AI agents and web applications from a single binary without managing Node.js or Python runtimes.
Enterprise teams looking for an all-in-one runtime that includes data modeling, REST APIs, UI rendering, and built-in search.

Key capabilities

Provides vector search with OpenAI or FastEmbed, knowledge graph retrieval, and hybrid GraphRAG.

Getting started

Download or run the single binary, then define data models in JSON or YAML. The setup is considered easy because the binary avoids complex environment configuration.

Evidence and sources

README: Yao is an open-source runtime for building AI agents and web applications — shipped as a single binary.
Release: Yao v1.0.0-rc6
README: Single Binary — Runs on ARM64/x64; no Python, Node, or containers needed on the host
README: Agent Framework
README: Vector Search — Embeddings with OpenAI or FastEmbed - **Knowledge Graph** — Entity-relationship retrieval - **GraphRAG** — Hybrid vector + graph search

Reviewed at: 2026-07-18T06:13:05Z