User Guide

This guide is for developers building applications on OmniRT. You'll find:

• Getting Started — install and first request
• Generation — text to image, image to image, text to video, image to video, talking head
• Serving — CLI, Python API, HTTP Server
• Deployment — CUDA, Ascend backend, domestic deployment, Docker
• Models — supported models, support status, roadmap
• Features — presets, validation, service schema, dispatch & queue, telemetry

Still new to OmniRT?

Start with Quickstart — one omnirt validate plus one omnirt generate — before diving deeper.

Looking for API reference?

A full auto-generated reference is in progress. For now the Python API guide covers generate, validate, pipeline, and every typed request helper.