Documentation for the Cognisoc family of projects.

A bare-metal C unikernel for serving large language models -- no OS, no overhead.

Run AI models directly on mobile devices. No cloud. No latency. Complete privacy.

Run local GGUF language models from .NET — one package, one format, one programming model.

Run any LLM locally. Use it from any language. Deploy anywhere.

A modular LLM inference runtime written in Rust.

Learn how LLMs work by building one in Zig -- from tensors to text generation.