SGLang Diffusion Cookbook

The diffusion cookbook is a set of model-specific, validated deployment guides. Each page should let a reader answer three questions quickly: what the model is good at, which checkpoint and base topology to choose, and which optional features change performance, memory, or quality.

What every model page includes

A compact capability tag row using the shared DiffusionModelTags component.
A short introduction that states the model's capability range, strongest use case, and an important limitation or deployment tradeoff.
Verified base launch recipes for the hardware actually tested.
Matching request examples for every public task or checkpoint partition.
Optional attention, quantization, caching, compile, and offload recipes with explicit quality contracts and validation scope.

Tags describe the model, not every runtime flag. Prefer video + audio, multi-reference, or 4–15 seconds over generic labels such as native, fast, or high quality.

Keep the command picker small

The command picker selects a runnable base recipe: hardware topology, checkpoint partition, request mode, and a validated placement profile. Independent feature knobs do not need to become picker dimensions.

Document orthogonal controls as feature overlays instead:

Feature class	Where it belongs	Required context
Attention backend	Optional feature overlays	Default backend, precision/approximation contract, installation, measured hardware
Online or pre-quantized weights	Optional feature overlays	Weight source, protected precision-sensitive layers, quality and memory tradeoff
Cache-DiT and sampling controls	Request/sampling recipes	Quality level, incompatible placement modes, task-specific validation
TP, Ulysses, Ring, FSDP, residency	Base picker only for verified cells	Exact GPU count, topology, memory, and latency evidence

For memory placement, prefer the unified --component-residency selector. Each component resolves to exactly one of resident, component-offload, or layerwise-offload. Existing options such as --dit-cpu-offload, --text-encoder-cpu-offload, --image-encoder-cpu-offload, and --vae-cpu-offload remain supported by all recipes that already use them.

Contributing a model

Use the repo-local cookbook-add-model skill and its templates/diffusion-page.mdx.tmpl template. The full category contract lives in .claude/skills/cookbook-add-model/references/diffusion-authoring.md.

Before opening a PR, run:

bash

node docs/scripts/check_cookbook_configs.mjs
cd docs
mint validate
mint broken-links --check-anchors --check-redirects

The static cookbook check requires every diffusion model page to use the shared tag widget, the standard introduction heading, 4–6 tags, and a substantive two-paragraph lead. Mintlify validation then catches MDX, navigation, and internal-link problems.

For runtime concepts and platform support, see the SGLang Diffusion documentation and compatibility matrix.