MODEL DEVELOPMENT / Hugging Face
Hugging Face LLM Trainer
Prepares and submits managed training jobs with dataset checks, monitoring and explicit persistence of trained artifacts.
“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 involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.”
01 / THE REASONING
Why this made the selection.
- Treats dataset shape, credentials and model persistence as prerequisites rather than leaving them until a job fails.
- Provides inspectable training and dataset-inspection scripts alongside reliability guidance.
02 / THE REVIEW RECORD
What we actually inspected.
Source review has boundaries.
A clear record is more useful than a “safe” badge.
Material inspected
- skills/huggingface-llm-trainer/SKILL.md
- skills/huggingface-llm-trainer/scripts/dataset_inspector.py
- skills/huggingface-llm-trainer/references/reliability_principles.md
- skills/huggingface-llm-trainer/scripts/train_sft_example.py
- README.md installation section
- LICENSE (Apache License 2.0 text)
Our findings
- The inspected training example creates a train/evaluation split, enables Trackio and pushes the resulting model to the Hub.
- The source warns that job storage is ephemeral and requires a write-capable token plus a chosen Hub destination.
- The reliability reference calls for checking actual model/dataset existence and running a small validation job before production work.
- Key Directives tell the agent to submit the job through the hf_jobs() MCP tool as soon as training is requested, to include Trackio in every script, and not to poll job status until the user asks.
- The frontmatter says "Complete terms in LICENSE.txt", but the skill directory holds no LICENSE.txt at this commit. The repository LICENSE is the Apache 2.0 text.
- Re-pinned from abc20ae526d8b4c0e4dff89f904adce28a4a0eb6 to c3ff942be3e07116b475253f0a8fc4f54682c1a8: a git tree comparison shows no change to any file under skills/huggingface-llm-trainer, to LICENSE or to README.md between the two commits. upstreamDescription now carries the full frontmatter description; the earlier quote stopped mid-sentence.
Not established by this review
- Source inspection only. No job was submitted, no script was run and the skill was not installed.
The review applies to the material and revision named here. A newer upstream release can change its behavior.
03 / PUT IT TO WORK
Add a skill to your workflow.
DOCUMENTED COMMAND
npx skills add huggingface/skills/skills/huggingface-llm-trainer Run in your terminal to choose your AI tools, project or global scope, and installation method. This short command follows the current upstream version; the review covers the revision shown above.
Optional: install the reviewed revision
Keeps the original source and installer version where recorded, while leaving your AI tools and installation scope for you to choose.
npx skills add https://github.com/huggingface/skills/tree/c3ff942be3e07116b475253f0a8fc4f54682c1a8/skills/huggingface-llm-trainer - Install the complete skill directory with its references and helpers; a copied SKILL.md alone may be incomplete.
- Choose your AI tools and project or global scope in the installer, then review its permissions before using the skill.
Before you start
- Node.js/npm for this installer
- An agent that supports the Agent Skills format
- Eligible paid Hugging Face plan and Jobs access
- Write-capable HF token, permitted data/model and chosen artifact repository
04 / FOLLOW THE EVIDENCE
The source trail.
Our notes are separate from the original resource.
Check upstream before adopting a new version.
https://github.com/huggingface/skills/blob/c3ff942be3e07116b475253f0a8fc4f54682c1a8/skills/huggingface-llm-trainer/SKILL.md
Supports: summary, upstreamDescription, whySelected, bestFor, limitations, review, compatibility, access
https://github.com/huggingface/skills/blob/c3ff942be3e07116b475253f0a8fc4f54682c1a8/skills/huggingface-llm-trainer/scripts/dataset_inspector.py
Supports: review, limitations, access
https://github.com/huggingface/skills/blob/c3ff942be3e07116b475253f0a8fc4f54682c1a8/skills/huggingface-llm-trainer/references/reliability_principles.md
Supports: review, limitations, access
https://github.com/huggingface/skills/blob/c3ff942be3e07116b475253f0a8fc4f54682c1a8/skills/huggingface-llm-trainer/scripts/train_sft_example.py
Supports: review, limitations, access
https://github.com/huggingface/skills/blob/c3ff942be3e07116b475253f0a8fc4f54682c1a8/LICENSE
Supports: license
https://github.com/huggingface/skills/blob/c3ff942be3e07116b475253f0a8fc4f54682c1a8/README.md
Supports: install, compatibility, access
https://github.com/vercel-labs/skills/blob/7407f3893ad4dceab546ac002c3ef806e4000c73/README.md
Supports: install
https://github.com/vercel-labs/skills/blob/7407f3893ad4dceab546ac002c3ef806e4000c73/src/git.ts
Supports: install