← All skills
tao-finetune-huggingface-model
80 / 100 · Low risk
Fine to try. Some of what it can do is expected: a coding skill usually runs build and test commands.
By NVIDIA/skills (GitHub) · coding · 3,933 stars · scanned at commit 2630133
Fine-tune any HuggingFace CV / VLM / LLM model on local NVIDIA GPUs inside an NGC PyTorch container when no dedicated TAO model skill matches. Use when the user wants to fine-tune a HuggingFace model (full or LoRA), train a vision / VLM / LLM model end-to-end, generate a reproducible HF training pipeline, smoke-test a HuggingFace model locally before scale-up, push a fine-tuned model to the HF Hub with a model card, or emit a self-contained rerun skill for an existing HuggingFace finetune. Supp…
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. Expected for a coding skill.
Found in 8 places.
Show 5 more
Downloads and installs other software
The author does not mention this.
Found in 15 places.
Show 12 more
- skills/tao-finetune-huggingface-model/references/deliverables.md, line 325
"${NGC_IMAGE}" -c "cd /workspace && pip install build -q && python -m build --wheel --outdir dist/" - skills/tao-finetune-huggingface-model/references/deliverables.md, line 357
docker exec "${DEV_CONTAINER}" bash -c "cd /workspace && pip install -e . -q" - skills/tao-finetune-huggingface-model/references/deliverables.md, line 470
**Why editable install in dev mode?** `pip install -e .` inside the container points entry - skills/tao-finetune-huggingface-model/references/hardware-container.md, line 161
bash -c "pip install transformers -q && python -c 'import transformers; print(transformers.__version__)'" - skills/tao-finetune-huggingface-model/references/pipeline-skill-template.md, line 130
all Python deps installed (no runtime `pip install`), so dropping root - skills/tao-finetune-huggingface-model/references/reporting.md, line 440
pip install matplotlib seaborn pillow pyyaml - skills/tao-finetune-huggingface-model/references/testing.md, line 497
"cd /workspace && pip install -r requirements.txt pytest -q && \ - skills/tao-finetune-huggingface-model/references/testing.md, line 498
pip install dist/*.whl -q 2>/dev/null || python -m build --wheel --outdir dist/ -q && pip install dist/*.whl -q && \ - skills/tao-finetune-huggingface-model/references/workflow-generate-train.md, line 113
RUN pip install --no-cache-dir -r requirements.txt - skills/tao-finetune-huggingface-model/references/workflow-intake-preflight.md, line 283
bash -c "pip install -q transformers huggingface_hub datasets Pillow && python model_probe.py" - skills/tao-finetune-huggingface-model/references/workflow-intake-preflight.md, line 290
- `HOME=/probe` + `PIP_USER=1` makes `pip install` resolve to - skills/tao-finetune-huggingface-model/references/workflow-intake-preflight.md, line 347
bash -c "pip install -q transformers huggingface_hub datasets Pillow && python dataset_probe.py"
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'tao-finetune-huggingface-model' ] || [ -L ~/.agents/skills/'tao-finetune-huggingface-model' ]; then echo 'A skill folder named tao-finetune-huggingface-model already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/NVIDIA/skills' "$d" && git -C "$d" checkout --quiet '2630133f22e357f2005886567a39eddcc1b520e2' -- 'skills/tao-finetune-huggingface-model' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/tao-finetune-huggingface-model' ~/.agents/skills/'tao-finetune-huggingface-model' && 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/'tao-finetune-huggingface-model' ~/.claude/skills/'tao-finetune-huggingface-model' (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/skills shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/NVIDIA/skills/)
Score history
| Date | Version | Score |
|---|
| 2026-11-11 | 2630133 | 80 |
| 2026-11-08 | fce3cca | 80 |
| 2026-11-07 | fce3cca | 80 |