Eino is ByteDance’s LLM application development framework for Go, part of the CloudWeGo ecosystem. It brings the component-and-orchestration approach popularized by LangChain to Go with strong typing and production-grade engineering, and it has collected roughly 13,400 stars on GitHub while powering AI features inside ByteDance products.
Go teams building LLM features often had to choose between Python frameworks and hand-rolled API calls. Eino closes that gap with reusable components for chat models, tools, retrievers, prompts and document handling, plus an orchestration layer that composes them into chains and graphs with compile-time type checking instead of runtime surprises. Streaming is a first-class concern throughout, so partial results flow through the graph as they are generated. Around the core, the EinoExt ecosystem provides integrations for model providers such as OpenAI, Claude, Gemini and Ollama, vector stores and other services, while companion tooling adds visual debugging and an IDE plugin for inspecting agent runs. Because it comes out of ByteDance’s internal platform work, the emphasis is on stability, observability and concurrency that behaves the way Go developers expect.
Eino is free and open source under the Apache-2.0 license. You only pay for the model APIs and infrastructure your application uses.
Eino is for Go backend teams building LLM applications, agents and RAG systems who want a structured, type-safe framework instead of ad-hoc API calls, especially teams already invested in the CloudWeGo microservices stack.