← All skills
deployment
77 / 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 NVIDIA/Model-Optimizer (GitHub) · devops · 5,579 stars · scanned at commit 6713c5a
Serve a quantized or unquantized LLM checkpoint as an OpenAI-compatible API endpoint using vLLM, SGLang, or TRT-LLM. Use when user says "deploy model", "serve model", "start vLLM server", "launch SGLang", "TRT-LLM deploy", "AutoDeploy", "benchmark throughput", "serve checkpoint", or needs an inference endpoint from a HuggingFace or ModelOpt-quantized checkpoint. Do NOT use for quantizing models (use ptq) or evaluating accuracy (use evaluation).
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
The author does not mention this. It runs automatically when the plugin is installed. Expected for a devops skill.
Found in 14 places.
Show 11 more
- plugins/modelopt/skills/deployment/SKILL.md, line 87
python -c "import vllm; print(f'vLLM {vllm.__version__}')" 2>/dev/null || echo "vLLM not installed" - plugins/modelopt/skills/deployment/SKILL.md, line 119
python -m vllm.entrypoints.openai.api_server \ - plugins/modelopt/skills/deployment/SKILL.md, line 148
python -m sglang.launch_server \ - plugins/modelopt/skills/deployment/SKILL.md, line 185
curl -s http://localhost:8000/health - plugins/modelopt/skills/deployment/SKILL.md, line 220
source "<common-skill-dir>/remote_exec.sh" - plugins/modelopt/skills/deployment/SKILL.md, line 231
remote_sync_to <local_checkpoint_path> <session_id>/<model>/checkpoints/ - plugins/modelopt/skills/deployment/SKILL.md, line 241
remote_run "nohup python -m vllm.entrypoints.openai.api_server --model <path> --port 8000 > deploy.log 2>&1 &" - plugins/modelopt/skills/deployment/SKILL.md, line 247
remote_run "curl -s http://localhost:8000/health" - plugins/modelopt/skills/deployment/references/setup.md, line 57
bash -c "python -m vllm.entrypoints.openai.api_server \ - plugins/modelopt/.claude-plugin/plugin.json, line 22
"command": "uvx", - plugins/modelopt/.mcp.json, line 3
"command": "uvx",
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 12 places.
Show 9 more
- plugins/modelopt/skills/deployment/SKILL.md, line 247
remote_run "curl -s http://localhost:8000/health" - plugins/modelopt/skills/deployment/SKILL.md, line 248
remote_run "curl -s http://localhost:8000/v1/models" - plugins/modelopt/skills/deployment/references/benchmarking.md, line 33
curl -s http://localhost:8000/v1/chat/completions \ - plugins/modelopt/skills/deployment/scripts/deploy.sh, line 362
if curl -s "http://localhost:$PORT/health" >/dev/null 2>&1; then - plugins/modelopt/skills/deployment/scripts/deploy.sh, line 430
if ! curl -s "http://localhost:$PORT/health" >/dev/null 2>&1; then - plugins/modelopt/skills/deployment/scripts/deploy.sh, line 438
curl -s "http://localhost:$PORT/v1/models" | python3 -m json.tool 2>/dev/null || true - plugins/modelopt/skills/deployment/scripts/deploy.sh, line 443
model_id=$(curl -s "http://localhost:$PORT/v1/models" | python3 -c " - plugins/modelopt/skills/deployment/scripts/deploy.sh, line 461
response=$(curl -s "http://localhost:$PORT/v1/completions" \ - plugins/modelopt/skills/deployment/tests/evals.json, line 36
"Verifies remotely: remote_run 'curl -s http://localhost:8000/health'",
Downloads and installs other software
The author does not mention this. It runs automatically when the plugin is installed. Expected for a devops skill.
Found in 10 places.
Show 7 more
- plugins/modelopt/skills/deployment/references/setup.md, line 32
pip install tensorrt-llm - plugins/modelopt/skills/deployment/references/sglang.md, line 29
- `pip install sglang[all]` - plugins/modelopt/skills/deployment/references/trtllm.md, line 7
- Or: `pip install tensorrt-llm` - plugins/modelopt/skills/deployment/references/unsupported-models.md, line 42
- **`VLLM_USE_PRECOMPILED=1 pip install --editable .`** — when patching vLLM source directly (instead of `sed`), this rebuilds only Python code without recompil - plugins/modelopt/skills/deployment/references/vllm.md, line 6
- `pip install vllm` - plugins/modelopt/.claude-plugin/plugin.json, line 22
"command": "uvx", - plugins/modelopt/.mcp.json, line 3
"command": "uvx",
Safer alternatives
Low-risk skills in the same category (devops) that look related. Check that one does what you need.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'deployment' ] || [ -L ~/.agents/skills/'deployment' ]; then echo 'A skill folder named deployment already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/NVIDIA/Model-Optimizer' "$d" && git -C "$d" checkout --quiet '6713c5a679fab80fdd9b688df01845f545c307b1' -- 'plugins/modelopt/skills/deployment' && mkdir -p ~/.agents/skills && cp -R "$d"/'plugins/modelopt/skills/deployment' ~/.agents/skills/'deployment' && rm -rf "$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/'deployment' ~/.claude/skills/'deployment' (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 NVIDIA/Model-Optimizer shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/NVIDIA/Model-Optimizer/)
Score history
| Date | Version | Score |
|---|
| 2026-11-10 | 6713c5a | 77 |