skills.rest
rayleizhu
rayleizhuCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

add-new-diffusion-model

Guide adding new diffusion models to the SGLang framework.

Updated Aug 23, 2026
Search Tags:#mlops#deep learning#diffusion models#model integration#sglang#pipeline configuration
One-click install
npx skills add https://github.com/rayleizhu/sglang --skill add-new-diffusion-model
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: add-new-diffusion-model
Source: https://github.com/rayleizhu/sglang/tree/main/python/sglang/multimodal_gen/.claude/skills/support-new-model
Command: npx skills add https://github.com/rayleizhu/sglang --skill add-new-diffusion-model

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and scripts (resource) components.

What problem does it solve?

This Skill provides a comprehensive guide and framework for integrating new diffusion models into the SGLang ecosystem, ensuring compatibility and efficient deployment.

Core Features & Use Cases

  • Model Integration: Step-by-step instructions for adding new diffusion models, supporting both Hybrid Monolithic and Modular Composition pipeline styles.
  • Pipeline Configuration: Guidance on setting up pipeline configurations, model components, registration, and testing procedures.
  • Use Case: A researcher has developed a novel diffusion model and needs to integrate it into SGLang for high-performance inference. This Skill provides the exact steps and code structure required for a successful integration.

Quick Start

Follow the tutorial to add a new diffusion model to SGLang.

Frequently Asked Questions about add-new-diffusion-model

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add a new diffusion model to SGLang for high-performance inference?▼

To add a diffusion model to SGLang, you must implement the model components, configure the pipeline, register the architecture, and verify the setup. This process supports both Hybrid Monolithic and Modular Composition pipeline styles for seamless integration.

What is the step-by-step process for integrating custom diffusion architectures into SGLang?▼

The integration process involves configuring the pipeline, implementing necessary model components, registering the new diffusion model, and finally verifying the setup. This ensures your custom architecture is fully compatible with the SGLang ecosystem.

Can I use a modular composition pipeline style when configuring diffusion models in SGLang?▼

Yes, SGLang supports the Modular Composition pipeline style for configuring diffusion models. This allows you to structure your model integration by setting up distinct, modular components within the pipeline configuration.

What are the limitations when adding novel diffusion architectures to SGLang?▼

Adding novel diffusion architectures requires strict adherence to SGLang's pipeline configuration and component registration rules. You must ensure your custom model components are correctly implemented and verified to avoid integration failures during deployment.

Related Skills

Explore More Software Engineering Skills→
guqiong96guqiong96
105

diffusion-optimal-perf

Optimize diffusion model inference with lossless and lossy techniques in SGLang-Diffusion.

Community
Advanced
scottgl9scottgl9
12

sglang-diffusion-modelopt-quant

Quantize diffusion transformers to FP8 or NVFP4 with NVIDIA ModelOpt.

Community
Advanced
hey-konghey-kong

sglang-diffusion-performance

Benchmark SGLang and Diffusers backends to optimize diffusion performance on multi-GPU setups.

Community
Advanced
sgl-projectsgl-project
31.6k

diffusion-kernel

Integrate custom JIT CUDA and Triton kernels to accelerate diffusion model inference.

Official
Advanced