lingoose

Why we included this project

Go developers who want to call LLMs from their existing services without switching languages will find LinGoose a compact, modular framework. It abstracts the pieces of an AI application, including model providers, embeddings, vector indexes, and conversation threads, behind Go interfaces, so you can swap implementations or plug in your own. The library takes you from a simple chat completion to retrieval-augmented pipelines backed by stores like Pinecone or Redis, which makes it handy for teams prototyping RAG features in Go. One honest caveat: the project is no longer under active development, and the author points new work toward a successor framework. Treat it as a stable, readable foundation for learning or for projects that can accept a maintained but frozen dependency, not something expecting ongoing feature growth.