documentation
Skill2.8k starsChanged 10 days agoApache-2.0View on GitHub
Update xDiT docs, examples, navigation, and diagrams against current source.
118 other projects use a version of this.
Trending
More talk lately.about the whole repository
Most recently taken up by
- arxdsilva/ai-loopstheir own version, first seen Nov 16, 2026
- AshrafDurjoy/genai-shared-patternstheir own version, first seen Nov 16, 2026
- th3r-dev/th3-skillstheir own version, first seen Nov 16, 2026
- cooneycw/claude-power-packtheir own version, first seen Nov 15, 2026
- lengrufeng2008/dsh-skill-librarytheir own version, first seen Nov 15, 2026
- scolladon/crafttheir own version, first seen Nov 15, 2026
Who's talking about xdit-project/xDiT
- zai-org/CogVideo“Collaboration with xDiT, where xDiT is a DiT parallel inference framework”GitHub · Aug 6, 2024 · 10 reactions
- Tencent-Hunyuan/HunyuanDiT“xDiT supports parallel inference of HunyuanDiT”GitHub · Aug 12, 2024 · 5 reactions
- @aime-hq.bsky.social“xDiT is a scalable inference Engine for diffusion transformers (DiTs). Think of it as vLLM for Pixels …”Bluesky · May 9, 2025 · 2 likes
- sgl-project/sglang“add fid accuracy benchmark for sglang diffusion t2i model”GitHub · May 20, 2026 · 1 reactions
- vllm-project/vllm-omni“[RFC]: Pipeline Parallelism & Stream Batch for Real-Time Video Generation”GitHub · Mar 27, 2026 · 1 reactions
The biggest 5 of 31 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 31
Talked about on GitHub and Bluesky (31 times).
On GitHub · 30
- Aug 6, 2024Collaboration with xDiT, where xDiT is a DiT parallel inference framework — zai-org/CogVideo10 reactions · 1 comment
- Aug 12, 2024xDiT supports parallel inference of HunyuanDiT — Tencent-Hunyuan/HunyuanDiT5 reactions
- May 20, 2026add fid accuracy benchmark for sglang diffusion t2i model — sgl-project/sglangsays it tested it1 reaction · 6 comments
- Mar 27, 2026[RFC]: Pipeline Parallelism & Stream Batch for Real-Time Video Generation — vllm-project/vllm-omni1 reaction · 5 comments
- Aug 26, 2025[RFC]: Supporting multi-modality models in AIBrix — vllm-project/aibrix1 reaction · 12 comments
25 more on GitHub
- Jun 6, 2025Add functional RingJointAttention op — tenstorrent/tt-metal1 reaction
- Oct 16, 2024[c10d] ProcessGroupNCCL support setting custom stream for communication — pytorch/pytorch1 reaction · 3 comments
- Aug 8, 2024Collaberation on parallel flux with xDiT — chengzeyi/piflux1 reaction · 2 comments
- Nov 10, 2026Enable new Quantized Attention A2A — AMD-AGI/diffusion-models-inference0 reactions
- Nov 2, 2026Install MORI — AMD-AGI/diffusion-models-inference0 reactions · 2 comments
- Oct 28, 2026[docker] bump diffusers — AMD-AGI/diffusion-models-inference0 reactions
- Oct 12, 2026xDiT: save capture trace for analysis — AMD-AGI/Magpie0 reactions
- Sep 23, 2026Feat/enable memory efficient loading generally — AMD-AGI/diffusion-models-inference0 reactions · 1 comment
- Sep 17, 2026[docker] bump to torch 2.13.0 — AMD-AGI/diffusion-models-inference0 reactions
- Sep 17, 2026fix: TorchAO re-quantization error — AMD-AGI/diffusion-models-inference0 reactions
- Sep 10, 2026Support for determinism check in xDiT — AMD-AGI/diffusion-models-inference0 reactions · 1 comment
- Aug 6, 2026Division by 0 bug when only one denoising step — huggingface/diffusers0 reactions
- Jul 21, 2026RDNA4 updates — AMD-AGI/diffusion-models-inference0 reactions · 1 comment
- May 26, 2026Quark xDiT integration and benchmark — ROCm/rocm-blogssays it tested it0 reactions
- Jan 21, 2026[New Feasture]: Support SP + DP parallal on Wan training — modelscope/DiffSynth-Studio0 reactions · 12 comments
- Dec 6, 2025[Feature] Support XDiT — Video-Reason/VBVR-EvalKit0 reactions · 3 comments
- Sep 3, 2025New HunyuanVideoI2V PR changes to transformer_hunyuan_video.py — huggingface/diffusers0 reactions · 5 comments
- Jun 18, 2025Mock policy distillation infrastructure — Overworldai/owl-wms0 reactions · 2 comments
- Mar 11, 2025multi-gpu support is long overdue — LykosAI/StabilityMatrix0 reactions · 2 comments
- Feb 21, 2025xdit multi gpu — kohya-ss/musubi-tuner0 reactions
- Feb 16, 2025Help , xdit support for multi gpu — kijai/ComfyUI-HunyuanVideoWrapper0 reactions · 4 comments
- Jan 20, 2025Awesome project and greeting from xDiT — zishen-ucap/LTX-Video-xDiT0 reactions · 1 comment
- Jan 9, 2025Enhancement: Implement DistVAE — leejet/stable-diffusion.cpp0 reactions · 1 comment
- Sep 4, 2024[docs] Add xDiT in section optimization — huggingface/diffusers0 reactions · 3 comments
- Aug 30, 2024Parallel Inference using xDiT — PKU-YuanGroup/Open-Sora-Plan0 reactions
On Bluesky · 1
- May 9, 2025xDiT is a scalable inference Engine for diffusion transformers (DiTs). Think of it as vLLM for Pixels (IMages, Video). xDiT is an inference… — @aime-hq.bsky.social2 likes and reposts
Used with
Projects with it also use create-media-pack, project-preferences and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
DepressionCenterSkillcreate-media-pack
Create a coordinated project branding and media pack with three accessible design options, approved vector artwork, image exports, a branding guide, and README integration. Use when asked for a repository media pack, brand kit, or complete project visual identity.
Also in 5 projects that use this
DepressionCenterSkillproject-preferences
Apply repository-specific preferences when planning, implementing, or reviewing changes in this project.
Also in 6 projects that use this
thistleknotSkillresponse-style
Response and prose protocol. Use when drafting user-facing prose, expanding a user's ideas, or tuning tone and format to preserve voice, avoid AI cliches, and maintain coherent answers.
Also in 7 projects that use this
BrowserWorksSkillaccessibility-frontend-review
Performs an accessibility code review of a local diff or Phabricator revision in the style of the Firefox accessibility team. Use this skill whenever the user asks to review a Phabricator patch or revision for accessibility issues, mentions a D-number (like D284142) in the context of accessibility review, asks "does this patch have any a11y issues", or wants to check a diff against accessibility guidelines. Also use it when the user says things like "can you review this for a11y", "check this patch", or "give me an accessibility review of D[number]". Invoke proactively when the user pastes a phabricator.services.mozilla.com URL or shares a revision ID alongside any accessibility concern. Invoke when questions about the following topics are raised: aria properties, aria roles, high contrast mode, forced colors, prefers contrast, focus order, screen readers, assistive technology.
Also in 4 projects that use this
BrowserWorksSkillandroid-new-module
Guide for creating new Android gradle modules in the android-components project.
Also in 4 projects that use this
- +7 more from BrowserWorks
Other versions118 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 118
- anthropics/knowledge-work-plugins★ 36ksame name
- comet-ml/opik★ 23ksame name
- mozilla-firefox/firefox★ 14ksame name
- microsoft/fast★ 9.7ksame name
- mufeedvh/code2prompt★ 7.8ksame name
- BrowserWorks/waterfox★ 6.4ksame name
- samchon/typia★ 6ksame name
- github/gh-aw★ 5.6ksame name
- epicenter-md/epicenter★ 4.8ksame name
- EpicenterHQ/epicenter★ 4.8ksame name
108 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
8 rules. 0 are common, 7 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 106 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use clear technical English inspired by ASD-STE100, without claiming formal compliance.
- Distinctive: line 11
Verify flags and features against the parser, adapters, backend specs, and execution.
- Distinctive: line 16
Check input/output metadata and the manual attention table against source.
- Distinctive: line 19
Use short, active sentences, consistent terms, and exact identifiers.
- Distinctive: line 20
Put prerequisites first and link to detail instead of repeating it.
- Distinctive: line 24
Update paired translations and check their links.
- Very specific: line 27
Run pre-commit run pymarkdown --all-files and python -m mkdocs build --strict.
- Distinctive: line 28
Check links, search, archives, and desktop/mobile layouts.
“Common” is measured: kept by at least 6 in 10 of the 106 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
xDiT documentation
Use clear technical English inspired by ASD-STE100, without claiming formal compliance.
- Verify flags and features against the parser, adapters, backend specs, and execution. Parser acceptance does not establish support.
- State inputs, defaults, dependencies, hardware, and restrictions. Check each example's model, flags, and process count. Label legacy interfaces.
- Regenerate CLI and model tables with
scripts/docs/; do not edit generated pages. Check input/output metadata and the manual attention table against source. - Preserve external sources' titles and verify dates. Keep descriptions factual. Distinguish published images from local container builds.
- Use short, active sentences, consistent terms, and exact identifiers. Put prerequisites first and link to detail instead of repeating it.
- Organize pages around user tasks. Keep general guidance vendor-neutral. Exclude edit history and review conversations.
- Archive retired guides and measurements in
docs/archive/. Update paired translations and check their links. - Preserve the official logo. Render accessible SVGs in both themes; check execution, text alignment, padding, and arrows.
- Run
pre-commit run pymarkdown --all-filesandpython -m mkdocs build --strict. Check links, search, archives, and desktop/mobile layouts. Keep flags intact and tables scrolling within the page.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

