YOUR HARDWARE. YOUR MODELS.

Powerful AI.
Closer to home.

Run open models on the Intel hardware you already own. From your first local request to a multi-machine inference pipeline.

Start building How it works

INTELLIGENCE, DISTRIBUTED.

✓ Open source✓ OpenAI-compatible✓ Built for IntelCheck compatibility

01 / GET STARTED

Your first step starts here.

One machine or many. Same starting point.

Make your first request

QUICKSTART

Download Cascadia, serve a ready-made model,
and send it a request. No build required.

Hello, Cascadia

Run your first model

GUIDE

Put your Intel hardware to work.
Export a model and serve it on one machine.

Set up single-machine inference

02 / EXPLORE THE CAPABILITIES

A small footprint. A bigger possibility.

SCALE OUT

Multi-machine inference

Run models larger than a single machine. Connect your Intel devices into a pipeline.

BUILD FASTER

OpenAI-compatible API

Keep your favorite tools and SDKs. Point your application at a local endpoint.

MAKE IT FIT

Built-in model sharding

Export and quantize Hugging Face models into ready-to-run INT4 stages.

USE YOUR HARDWARE

Intel-native engines

Choose the right engine for your model, from OpenVINO to specialized MoE paths.

STAY INFORMED

Visibility at every stage

Track token throughput, latency, and pipeline health with Prometheus metrics.

STAY IN CONTROL

Inference on your terms

Keep inference on your own network. Understand the deployment security model.

Go a level deeper.

The commands, internals, and interfaces behind Cascadia.

CLI reference HTTP API