DevOnBike 

Security powered by Heimdall

Pure-C# in-process LLM and deep-learning runtime for .NET 10. Loads GGUF / HF safetensors / ONNX directly (Qwen, Llama, Mistral, Mixtral, MoE, BERT-family embeddings). Zero-allocation CPU inference, Native... More information

c# functional extensions powered by DevOnBike

Use the in-process Overfit LLM runtime through Microsoft.Extensions.AI. Implements IChatClient and IEmbeddingGenerator, so it plugs into Semantic Kernel and any Microsoft.Extensions.AI pipeline — function... More information

A dependency-free, OpenAI-compatible HTTP server for the in-process Overfit LLM runtime — built on System.Net.HttpListener + System.Text.Json source-gen, no ASP.NET Core, Native-AOT friendly. Point any OpenAI... More information

Overfit CLI — a pure-.NET, OpenAI-compatible local LLM server (`overfit serve`) plus model pull/chat/embed/tts. No Python, no native engine. Also distributed as a Native-AOT binary and a Docker image.

Package Description

A dependency-free MCP (Model Context Protocol) stdio server for the in-process Overfit LLM runtime — typed JSON-RPC 2.0 contracts with source-generated System.Text.Json serialization, no SDK dependency, no... More information