# Cascadia > Cascadia is an open-source (Apache-2.0) inference runtime that serves open LLMs on Intel hardware (AI PCs, Arc GPUs, NPUs) through an OpenAI-compatible HTTP API, and can split one model across several machines as a pipeline. - The HTTP API and inter-stage relay have no TLS or authentication; deploy on a trusted LAN. - `cascadia run` and `cascadia worker` load local model directories only; they never download models. `cascadia shard` exports Hugging Face models into per-stage OpenVINO IR. - Every page is also available as Markdown by appending `.md` to its URL, e.g. https://docs.cascadia.to/getting-started/quickstart.md ## Documentation Sets - [Abridged documentation](https://docs.cascadia.to/llms-small.txt): a compact version of the documentation for Cascadia, with non-essential content removed - [Complete documentation](https://docs.cascadia.to/llms-full.txt): the full documentation for Cascadia - [Start here](https://docs.cascadia.to/_llms-txt/start-here.txt): quickstart, installation, and hardware and OS requirements - [Inference](https://docs.cascadia.to/_llms-txt/inference.txt): exporting models, single- and multi-machine inference, making requests to the OpenAI-compatible API - [Operations](https://docs.cascadia.to/_llms-txt/operations.txt): running workers as services, metrics, security model - [Reference](https://docs.cascadia.to/_llms-txt/reference.txt): HTTP API, CLI commands, flags, and inference engines, troubleshooting, Rust crates - [Enterprise](https://docs.cascadia.to/_llms-txt/enterprise.txt): the hosted console, adding devices, the Fleet Manager, API tokens, security, FAQ and the edition comparison ## Notes - The complete documentation includes all content from the official documentation - The content is automatically generated from the same source as the official documentation ## Optional - [Source code](https://github.com/labscommunity/cascadia): Repository, including in-depth engineering notes under docs/