vllm-projectOfficial·9 Agent Skills Included

vllm-omni

Serve multimodal AI models fast with guided testing, review, and optimization

Guides AI agents through building, testing, and optimizing vLLM-Omni, a serving framework for text, image, audio, and video models. Covers adding diffusion and TTS models, writing CI-aligned tests, reviewing pull requests, applying quantization, and tuning GPU performance. Removes guesswork from complex model-serving workflows with step-by-step checklists and proven code patterns.

6,536|1,594

↗

npx skills add vllm-project/vllm-omni --all -g -y

All Skills in This Repository (9)

Pure Emerald Level Indicators

📦 In Repo

vllm-project

add-diffusion-model

Add diffusion models to vLLM-Omni with Cache-DiT acceleration and multi-GPU parallelism.

vllm-omni-npu-model-runner-upgrade

Upgrade vllm-omni NPU model runners to align with vllm-ascend interfaces.

add-tts-model

Integrate HuggingFace TTS models into vLLM-Omni with CUDA graph acceleration.

vllm-omni-test

Generates and runs pytest tests for vllm-omni with CI-aligned levels, markers, and Buildkite wiring.

precheck-pr

Reviews branch diffs against vllm-omni contribution checklists before opening a pull request.

quantization

Configure and validate quantization methods for vLLM-Omni diffusion, autoregressive, and omni models.

find-simplifications

Audit vLLM-Omni code for evidence-backed simplification and deletion candidates.

diffusion-perf-opt

Diagnose and optimize vLLM Omni diffusion workloads using profiler traces and parallel strategy analysis.

review-pr

Review vLLM-Omni pull requests with frozen snapshots, module contracts, and evidence-backed findings.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install vllm-omni skills?▼

Run `npx skills add vllm-project/vllm-omni --all -g -y` in your terminal to install all skills in this suite globally.

What is vLLM-Omni used for?▼

vLLM-Omni is a serving framework for omni-modality AI models that handle text, image, audio, video, and robot actions. These skills help your AI agent test, review, and optimize that framework.

How do I add a new diffusion model to vLLM-Omni?▼

The add-diffusion-model skill walks your agent through adapting transformers, registering pipelines, enabling Cache-DiT acceleration, and wiring up tests and documentation.

Can these skills help review pull requests?▼

Yes. The review-pr and precheck-pr skills perform maintainer-grade reviews and pre-submit self-checks with evidence-backed findings and clear severity ratings.

Do these skills work with Claude Code and other AI agents?▼

Yes. All skills follow the universal SKILL.md standard and run in Claude Code, OpenClaw, Cursor, and Codex.