Overfit
A .NET library and CLI that runs, fine-tunes and serves large language models in-process, adding local chat, RAG, agents and voice to .NET applications without Python, a GPU or a separate model server.
Overfit brings local LLM inference, retrieval-augmented generation, agents and voice into .NET applications with no Python, external model server or cloud dependency. It implements Microsoft.Extensions.AI's IChatClient and IEmbeddingGenerator interfaces, so a local model can be swapped into the standard .NET AI template or Semantic Kernel by changing one line.
The library serves an OpenAI-compatible API from a small Native-AOT binary or a roughly 46 MB container, and separates a reasoning model's chain-of-thought from its final answer. It also supports CPU-only QLoRA fine-tuning on a developer's own data.
It installs as a NuGet package, a global CLI tool, and a set of project templates for ASP.NET, WPF, Blazor and desktop apps, aimed at regulated or offline environments where external LLM APIs are restricted.