Skill

vllm-ascend-model-adapter

by vllm-project

AI Summary

A skill for adapting and optimizing Hugging Face or custom LLM models to run efficiently on vLLM with Ascend NPU support, enabling developers to validate and deploy models with deterministic testing and single-commit delivery.

Description

Adapt and debug existing or new models for vLLM on Ascend NPU. Implement in /vllm-workspace/vllm and /vllm-workspace/vllm-ascend, validate via direct vllm serve from /workspace, and deliver one signed commit in the current repo.

Install

# Add to your project root as SKILL.md
curl -o SKILL.md "https://raw.githubusercontent.com/vllm-project/vllm-ascend/main/.agents/skills/vllm-ascend-model-adapter/SKILL.md"

Quality Score

B

Good

80/100

Standard Compliance72
Documentation Quality65
Usefulness78
Maintenance Signal100
Community Signal100
Scored Today

GitHub Signals

Stars1.7k
Forks853
Issues1263
UpdatedYesterday
View on GitHub

Trust & Transparency

Open Source — Apache-2.0

Source code publicly auditable

Verified Open Source

Hosted on GitHub — publicly auditable

Actively Maintained

Last commit Yesterday

1.7k stars — Strong Community

853 forks

My Fox Den

Community Rating

Works With

Claude Code