← All skills
vllm-ascend-server
79 / 100 · Review before installing
Check before installing. Some of what it can do is expected: a devops skill usually runs commands, installs tools and uses cloud credentials.
By ascend-ai-coding/awesome-ascend-skills (GitHub) · devops · 184 stars · scanned at commit 75ec89c
Deploy vLLM inference services on Ascend NPU servers with automatic model detection and optimized configuration. Supports local and remote deployment across bare metal, containers, and Docker images. Handles model discovery, quantization auto-detection, tensor parallelism configuration, graph/eager mode selection, and service health verification. Use when users need to: (1) Start or deploy vLLM server on NPU, (2) Launch LLM inference service, (3) Configure multi-card tensor parallel deployment,…
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
This is disclosed by the author in the skill's SKILL.md or README. Expected for a devops skill.
Found in 22 places.
Show 17 more
- skills/inference/vllm-ascend-server/SKILL.md, line 133
/home/weights, /home/weight, /home/data*, /data* - skills/inference/vllm-ascend-server/SKILL.md, line 142
[ -f "<model>/quant_model_description.json" ] → --quantization ascend - skills/inference/vllm-ascend-server/SKILL.md, line 188
export TASK_QUEUE_ENABLE=1 - skills/inference/vllm-ascend-server/SKILL.md, line 196
export VLLM_USE_V1=1 - skills/inference/vllm-ascend-server/SKILL.md, line 206
vllm serve /model/<model-name> \ - skills/inference/vllm-ascend-server/SKILL.md, line 248
nohup vllm serve /model ... > /tmp/vllm.log 2>&1 & - skills/inference/vllm-ascend-server/SKILL.md, line 251
docker exec -d <container> bash -c 'cd /workspace && vllm serve /model ... 2>&1 | tee /tmp/vllm.log' - skills/inference/vllm-ascend-server/SKILL.md, line 254
ssh user@host "docker exec -d <container> bash -c 'vllm serve /model ... 2>&1 | tee /tmp/vllm.log'" - skills/inference/vllm-ascend-server/SKILL.md, line 262
ps aux | grep vllm - skills/inference/vllm-ascend-server/SKILL.md, line 270
docker exec <container> ps aux | grep vllm - skills/inference/vllm-ascend-server/SKILL.md, line 279
curl http://localhost:8000/health - skills/inference/vllm-ascend-server/SKILL.md, line 292
export TASK_QUEUE_ENABLE=1 - skills/inference/vllm-ascend-server/SKILL.md, line 307
vllm serve /model/Qwen3-8B \ - skills/inference/vllm-ascend-server/SKILL.md, line 318
export VLLM_USE_V1=1 - skills/inference/vllm-ascend-server/SKILL.md, line 333
docker run -it -d \ - skills/inference/vllm-ascend-server/references/launch-templates/docker.md, line 37
bash -c " - skills/inference/vllm-ascend-server/references/launch-templates/docker.md, line 94
docker exec -d vllm-server bash -c 'vllm serve /model --port 8000 --quantization ascend'
and 2 more places
Connects to the internet
This is disclosed by the author in the skill's SKILL.md or README. Expected for a devops skill.
Found in 13 places.
Show 10 more
- skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 16
curl http://localhost:8000/v1/models - skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 26
curl http://localhost:8000/v1/chat/completions \ - skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 40
curl http://localhost:8000/v1/completions \ - skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 52
curl http://localhost:8000/v1/chat/completions \ - skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 84
curl http://<server-ip>:8000/health - skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 87
ssh user@host "curl -s localhost:8000/health" - skills/inference/vllm-ascend-server/references/launch-templates/health-check.md, line 98
response = requests.get("http://localhost:8000/health") - skills/inference/vllm-ascend-server/references/workflow-guide.md, line 409
curl http://localhost:8000/health - skills/inference/vllm-ascend-server/references/workflow-guide.md, line 417
curl http://localhost:8000/v1/models - skills/inference/vllm-ascend-server/references/workflow-guide.md, line 423
curl http://localhost:8000/v1/chat/completions \
Downloads and installs other software
The author does not mention this. Expected for a devops skill.
Found in 2 places.
Safer alternatives
Low-risk skills in the same category (devops) that look related. Check that one does what you need.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'vllm-ascend-server' ] || [ -L ~/.agents/skills/'vllm-ascend-server' ]; then echo 'A skill folder named vllm-ascend-server already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/ascend-ai-coding/awesome-ascend-skills' "$d" && git -C "$d" checkout --quiet '75ec89ce9d401be8ac95980e1c3f42f8d4315057' -- 'skills/inference/vllm-ascend-server' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/inference/vllm-ascend-server' ~/.agents/skills/'vllm-ascend-server' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'vllm-ascend-server' ~/.claude/skills/'vllm-ascend-server' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for ascend-ai-coding/awesome-ascend-skills shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/ascend-ai-coding/awesome-ascend-skills/)
Score history
| Date | Version | Score |
|---|
| 2026-11-08 | 75ec89c | 100 |