hsliuustc0106Community·14 Agent Skills Included

vllm-omni-skills

Deploy, tune, and review vLLM-Omni multimodal inference

Guides installation, serving, and scaling of vLLM-Omni for text, image, video, and audio model inference. Covers quantization, performance tuning, distributed multi-GPU setups, hardware backends, and CI/CD pipelines. Automates PR reviews, release notes, nightly test reports, and new model integration workflows. Eliminates manual configuration errors and slow trial-and-error when deploying omni-modality models.

84|27

↗

npx skills add hsliuustc0106/vllm-omni-skills --all -g -y

Available:

Tells the agent how this skill collection is structured, how to route requests to the right vLLM-Omni skill, and what naming, validation, and versioning rules to follow when working with the skills.

All Skills in This Repository (14)

Pure Emerald Level Indicators

📦 In Repo

hsliuustc0106

vllm-omni-audio-tts

Generate speech and audio with vLLM-Omni using Qwen3-TTS, MiMo-Audio, and Stable-Audio.

vllm-omni-video-gen

Generate videos from text and images using vLLM-Omni with Wan2.2 models.

vllm-omni-image-gen

Generate and edit images from text prompts using vLLM-Omni diffusion models.

vllm-omni-setup

Install and configure vLLM-Omni with Python environments and smoke tests.

vllm-omni-serving

Launch and configure vLLM-Omni API servers for OpenAI-compatible models.

vllm-omni-multimodal

Orchestrate multimodal reasoning and generation across text, image, audio, and video inputs.

vllm-omni-hardware

Configure vLLM-Omni hardware backends across CUDA, ROCm, NPU, and XPU.

vllm-omni-cicd

Automate CI/CD pipelines for vLLM-Omni deployments with Docker builds and GitHub Actions.

vllm-omni-api

Bridge client applications with the vLLM-Omni API using OpenAI-compatible endpoints.

vllm-omni-quantization

Quantizes vLLM-Omni models using AWQ, GPTQ, or FP8 for reduced memory and faster throughput.

vllm-omni-review

Review vllm-omni PRs for code quality, tests, and security issues.

vllm-omni-perf

Identify and optimize vLLM-Omni performance bottlenecks across autoregressive and diffusion pipelines.

Community

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install vllm-omni-skills?▼

Run `npx skills add hsliuustc0106/vllm-omni-skills --all -g -y` in your terminal to install all skills globally.

What does vllm-omni-skills do?▼

It gives your agent step-by-step workflows for installing, serving, quantizing, and scaling vLLM-Omni models across text, image, video, and audio. It also covers PR reviews, release notes, CI/CD, and nightly test reporting.

Which models and hardware are supported?▼

Skills cover Qwen-Omni, Qwen-Image, FLUX, Wan2.2 video, Qwen3-TTS, Fish Speech, and more. Hardware guidance spans NVIDIA CUDA, AMD ROCm, Huawei NPU, and Intel XPU.

Does it work with Claude Code and Cursor?▼

Yes. All skills follow the standard SKILL.md format and run in Claude Code, Cursor, Codex, and other compatible agents.

Can it help review vLLM-Omni pull requests?▼

Yes. The review skills route PRs to domain checklists, scan for blocking issues, verify performance claims, and can even mine historical PR reviews to build new review skills.