Agent Native

huggingface-llm-trainer

Train language models on Hugging Face cloud GPUs using TRL with methods including SFT, DPO, GRPO, and reward modeling.

by Hugging FaceOfficial publisherNot reviewed

Identity proven by GitHub's verified organisation badge (huggingface)

Agents
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode
License
License unclear
Pin
c3ff942pinned Oct 9, 2026
Context cost
about 7,000 tokenswhen the skill loads
GitHub stars
11,150on huggingface/skills
Scripts
8 scripts

What it does

“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.”
The author's description, from SKILL.md

Generated summary from the skill's own text

Train language models on Hugging Face cloud GPUs using TRL with methods including SFT, DPO, GRPO, and reward modeling. Submit jobs via the hf_jobs MCP tool, monitor with Trackio, and save results to the Hub. Includes GGUF conversion for local deployment and cost estimation tools.

When to use it

  • Fine-tune models on cloud GPUs without local infrastructure
  • Train using TRL methods (SFT, DPO, GRPO) for instruction tuning or alignment
  • Convert trained models to GGUF for Ollama, LM Studio, or llama.cpp
  • Estimate training costs and select appropriate hardware

When not to

  • User has local GPU and prefers training locally
  • Training only requires inference, not model fine-tuning
  • Dataset format validation using other tools would be faster

Install

Restricted license: install from the publisher's repository and read its terms.

These commands install from the publisher's repository and need no account.

skills CLI

npx skills add huggingface/skills --skill huggingface-llm-trainer

Installs from the repository's default branch, which may have changed since the commit we checked.

GitHub CLI 2.90 or later

gh skill install huggingface/skills huggingface-llm-trainer@c3ff942be3e07116b475253f0a8fc4f54682c1a8

Installs the commit we checked (c3ff942).

Copy by hand

Download the folder at c3ff942 and put it in .claude/skills/huggingface-llm-trainer for Claude Code and in .agents/skills/huggingface-llm-trainer for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.

Skill folder on GitHub at c3ff942

Claude Code plugin

/plugin install huggingface-skills@claude-plugins-official

The publisher's plugin in Claude's official marketplace. It may hold more than this skill, and we have not pinned it.

Copy for your agent

A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.

Prompt

Install the agent skill "huggingface-llm-trainer" by Hugging Face.

Source, pinned to commit c3ff942be3e07116b475253f0a8fc4f54682c1a8:
https://github.com/huggingface/skills/tree/c3ff942be3e07116b475253f0a8fc4f54682c1a8/skills/huggingface-llm-trainer

1. Download that folder at exactly this commit. Do not use the default branch.
2. Copy every file unchanged, keeping the folder structure, into .claude/skills/huggingface-llm-trainer/ for Claude Code and .agents/skills/huggingface-llm-trainer/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist.
3. Do not edit the files, and do not run any script in the folder unless I ask you to.
4. When you are done, list the files you copied and where you put them.

Its license is restricted or not stated. Show me the license terms from the repository before you copy anything.

Listing and checks: https://www.agentnative.dev/skills/hugging-face/huggingface-llm-trainer

SKILL.md at the pin

The license of this skill does not let us show its files here.

View SKILL.md on GitHub at c3ff942

Files

19 files, 186 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.

Skill folder on GitHub at c3ff942
All 40 skills from Hugging Face

More from Hugging Face

  • hf-cloud-aws-context-discovery

    This skill reads the user's local AWS configuration to discover their active profile, region, account ID, and caller identity before any AWS work begins.

    Claude Code, Codex, Cursor +4 · about 830 tokens · pinned 6 days ago

  • hf-cloud-python-env-setup

    Sets up an isolated Python environment for SageMaker and AWS work with the correct Python version (3.10, 3.11, or 3.12) and current boto3 and awscli.

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 1.5K tokens · pinned 6 days ago

  • hf-cloud-sagemaker-deployment-planner

    This skill helps users plan and execute deployments of machine learning models to Amazon SageMaker by gathering requirements, selecting appropriate deployment…

    Claude Code, Codex, Cursor +4 · about 1.9K tokens · pinned 6 days ago

  • hf-cloud-sagemaker-iam-preflight

    This skill validates or creates an AWS IAM execution role before deploying SageMaker endpoints, training jobs, or models.

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 1.6K tokens · pinned 6 days ago

  • hf-cloud-sagemaker-production-defaults

    Deploys SageMaker endpoints with production defaults: autoscaling based on traffic, CloudWatch alarms for latency and errors, and consistent tagging.

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 6.7K tokens · pinned 6 days ago

  • hf-cloud-serving-image-selection

    This skill helps select the correct SageMaker serving container image for model deployments and retrieves its current URI from AWS's Deep Learning Containers…

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 4.3K tokens · pinned 6 days ago

All Cloud and DevOps skills

Related in Cloud and DevOps

  • hf-mem

    This skill estimates memory requirements for loading machine learning models from Hugging Face Hub for inference, supporting both Safetensors and GGUF formats.

    by Hugging FaceOfficial publisher

    Claude Code, Codex, Cursor +4 · about 790 tokens · pinned 6 days ago

  • huggingface-lora-space-builder

    This skill builds and publishes a Gradio demo on Hugging Face Spaces that runs inference with a user-provided LoRA model.

    by Hugging FaceOfficial publisher

    Claude Code, Codex, Cursor +4 · about 8.2K tokens · pinned 6 days ago

  • huggingface-spaces

    Create and deploy machine-learning applications on Hugging Face Spaces using Gradio, Docker, or Static SDKs.

    by Hugging FaceOfficial publisher

    Claude Code, Codex, Cursor +4 · about 4.3K tokens · pinned 6 days ago

  • huggingface-vision-trainer

    Trains vision models for object detection, image classification, and SAM/SAM2 segmentation on Hugging Face Jobs cloud GPUs.

    by Hugging FaceOfficial publisher

    Contains scripts

    Claude Code · about 7.3K tokens · pinned 6 days ago

  • huggingface-zerogpu

    This skill provides rules and patterns for building ML demos on Hugging Face Spaces with ZeroGPU hardware, covering the @spaces.GPU decorator, duration and…

    by Hugging FaceOfficial publisher

    Claude Code, Codex, Cursor +4 · about 4.3K tokens · pinned 6 days ago

  • huggingface-spaces

    This skill enables agents to create, deploy, and troubleshoot machine learning applications on Hugging Face Spaces using Gradio, Docker, or Static SDKs.

    by Hugging FaceOfficial publisher

    Claude Code, Codex, Cursor +4 · about 4K tokens · pinned 6 days ago

Open roles at Hugging Face

3 open roles on the jobs board

See all 3 roles at Hugging Face on the jobs board

The skills listed here are third-party software, collected from public GitHub repositories as a catalog of recent skills and provided as is. Agent Native does not write or control them and is not responsible for any harm, loss or damage caused by installing or using them. Labels and checks describe what we saw at one commit and are not a guarantee. Read a skill's files before you run it. Terms of use