add-diffusion-model

Official

Add diffusion models to vLLM-Omni with speed.

Authorvllm-project
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This skill reduces the complexity of adding a new diffusion model to vLLM-Omni by providing a repeatable workflow that handles compatibility, acceleration, and multi-GPU deployment.

Core Features & Use Cases

  • Supports both Diffusers-based pipelines and custom repositories.
  • Guides porting to vLLM-Omni transformers and integrating Cache-DiT acceleration.
  • Covers registration, testing, documentation, and offline download script scaffolding.
  • Enables multi-GPU parallelism patterns (TP, SP/USP, CFG, HSDP) and production-ready workflows.

Quick Start

Create or adapt a diffusion model, register it in the diffusion registry, and run an offline test script to verify basic functionality.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: add-diffusion-model
Download link: https://github.com/vllm-project/vllm-omni/archive/main.zip#add-diffusion-model

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.