# ElBruno.LocalLLMs

A C# library that runs local large language models through the same IChatClient interface used for Azure OpenAI or Ollama, powered by ONNX Runtime GenAI, with Blazor components for building local-model chat interfaces.

ElBruno.LocalLLMs lets a .NET application run local language models through IChatClient, the same abstraction used with Azure OpenAI or Ollama, so switching between a hosted and a local model is a configuration change rather than a rewrite. It is powered by ONNX Runtime GenAI and BitNet, downloads models automatically from Hugging Face on first use, and works across CPU, CUDA, and DirectML execution providers with a default of a small Phi model.

A companion Razor Class Library adds ready-to-use Blazor components, including a model status card with download progress, a filterable model gallery, a model selector, a streaming chat box, an environment dashboard showing available hardware, and a health-check badge. Recent releases added support for OpenAI's open-weight GPT-OSS 20B model, including its Harmony prompt format and tool calling.

Website: https://github.com/elbruno/ElBruno.LocalLLMs