← All skills
huggingface-llm-trainer
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 huggingface/skills (GitHub) · devops · 11,321 stars · scanned at commit 25910b3
Train or fine-tune language and vision models using TRL (Transformer Reinforcement Learning) or Unsloth with Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts with PEP 723 format, dataset preparation and validation, hardware selection, cost estimation, Trackio monitoring, Hub authentication, model selection/leaderboards and model persistence. Use for tasks…
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 30 places.
Show 17 more
- skills/huggingface-llm-trainer/SKILL.md, line 313
hf jobs uv run \ - skills/huggingface-llm-trainer/SKILL.md, line 322
hf jobs ps # List all jobs - skills/huggingface-llm-trainer/SKILL.md, line 333
uvx trl-jobs sft \ - skills/huggingface-llm-trainer/SKILL.md, line 431
uv run scripts/hf_benchmarks.py --help - skills/huggingface-llm-trainer/SKILL.md, line 437
uv run scripts/hf_benchmarks.py search --query ocr - skills/huggingface-llm-trainer/SKILL.md, line 448
uv run scripts/estimate_cost.py \ - skills/huggingface-llm-trainer/SKILL.md, line 647
uv run https://huggingface.co/datasets/mcp-tools/skills/raw/main/dataset_inspector.py \ - skills/huggingface-llm-trainer/references/gguf_conversion.md, line 31
subprocess.run(["apt-get", "update", "-qq"], check=True, capture_output=True) - skills/huggingface-llm-trainer/references/gguf_conversion.md, line 32
subprocess.run(["apt-get", "install", "-y", "-qq", "build-essential", "cmake"], check=True, capture_output=True) - skills/huggingface-llm-trainer/references/gguf_conversion.md, line 44
subprocess.run([ - skills/huggingface-llm-trainer/references/gguf_conversion.md, line 50
subprocess.run([ - skills/huggingface-llm-trainer/references/reliability_principles.md, line 100
subprocess.run(["make", "-C", "/tmp/llama.cpp", "llama-quantize"], check=True) - skills/huggingface-llm-trainer/references/reliability_principles.md, line 103
subprocess.run([ - skills/huggingface-llm-trainer/references/reliability_principles.md, line 108
subprocess.run([ - skills/huggingface-llm-trainer/references/reliability_principles.md, line 183
subprocess.run(["git", "clone", "https://github.com/ggerganov/llama.cpp.git", "/tmp/llama.cpp"]) - skills/huggingface-llm-trainer/references/reliability_principles.md, line 184
subprocess.run(["make", "-C", "/tmp/llama.cpp", "llama-quantize"]) # FAILS: no gcc/make - skills/huggingface-llm-trainer/references/reliability_principles.md, line 187
subprocess.run(["apt-get", "update", "-qq"], check=True)
and 10 more places
Downloads and installs other software
The author does not mention this. Expected for a devops skill.
Found in 12 places.
Show 9 more
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 6 places.
Show 3 more
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/'huggingface-llm-trainer' ] || [ -L ~/.agents/skills/'huggingface-llm-trainer' ]; then echo 'A skill folder named huggingface-llm-trainer already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/huggingface/skills' "$d" && git -C "$d" checkout --quiet '25910b3d0ba31a2ba102f901cdc9e008c278a69c' -- 'skills/huggingface-llm-trainer' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/huggingface-llm-trainer' ~/.agents/skills/'huggingface-llm-trainer' && 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/'huggingface-llm-trainer' ~/.claude/skills/'huggingface-llm-trainer' (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 huggingface/skills shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/huggingface/skills/)
Score history
| Date | Version | Score |
|---|
| 2026-11-08 | 25910b3 | 79 |
| 2026-11-06 | 25910b3 | 59 |