hugging-face-model-trainer
Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.
npx skills add majiayu000/claude-skill-registry --skill hugging-face-model-trainer-sickn33-antigravity-awesome --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Instructs the agent to train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export, using cloud GPUs and pushing results to the Hub.
How it works
- Requires submission of training jobs via the hf_jobs() MCP tool with an inline script that defines the training configuration (SFT, DPO, GRPO, etc.) and model details.
- Each training script must include Trackio for real-time monitoring and push results to the Hugging Face Hub.
- Provide job details after submission: Job ID, monitoring URL, estimated time, and a Trackio dashboard URL.
- Use example templates from scripts/train_sft_example.py, scripts/train_dpo_example.py, etc., as starting points.
- Inline dependencies and scripts are preferred; local files are not saved unless requested.
When to use it
- When users want to fine-tune language models on cloud GPUs without local infrastructure
- When training with TRL methods (SFT, DPO, GRPO, etc.)
- When converting trained models to GGUF for local deployment and ensuring hub persistence
What it can touch
- Tools: hf_jobs() MCP tool
- Script content: inline Python code for TRL training (SFT/DPO/GRPO/Reward Modeling)
- Hub access: requires HF_TOKEN for hub push (passed via secrets)
Caveats
- Risk is listed as unknown
- Ephemeral environment: training results must be pushed to Hub to avoid loss
- Timeout settings are critical; default 30 minutes is insufficient for most training
- Requires HF_ACCOUNT with appropriate plan and write access to hub_model_id
# TRL Training on Hugging Face Jobs ## Overview Train language models using TRL (Transformer Reinforcement Learning) on fully managed Hugging Face infrastructure. No local GPU setup required—models train on cloud GPUs and results are automatically saved to the Hugging Face Hub. **TRL provides multiple training methods:** - **SFT** (Supervised Fine-Tuning) - Standard instruction tuning - **DPO** (Direct Preference Optimization) - Alignment from preference data - **GRPO** (Group Relative Policy Optimization) - Online RL training - **Reward Modeling** - Train reward models for RLHF **For detailed TRL method documentation:** ```python hf_doc_search("your query", product="trl") hf_doc_fetch("https://huggingface.co/docs/trl/sft_trainer") # SFT hf_doc_fetch("https://huggingface.co/docs/trl/dpo_trainer") # DPO # etc. ``` **See also:** `references/training_methods.md` for method overviews and selection guidance ## When to Use This Skill Use this skill when users want to: - Fine-tune language models on cloud GPUs without local infrastructure - Train with TRL methods (SFT, DPO, GRPO, etc.) - Run training jobs on Hugging Face Jobs infrastructure - Convert trained models to GGUF for local deplo
- Overview
- When to Use This Skill
- When to Use Unsloth
- Key Directives
- Local Script Execution
- Prerequisites Checklist
- ✅ Account & Authentication
- ✅ Dataset Requirements
- ⚠️ Critical Settings
- Asynchronous Job Guidelines
- Action Required
- Ground Rules
- After Submission
- Quick Start: Three Approaches
uv run scripts/estimate_cost.py --help uv run scripts/dataset_inspector.py --help hf repos create my-training-scripts --type model hf upload my-training-scripts ./train.py train.py hf jobs uv run --flavor a10g-large --timeout 2h --secrets HF_TOKEN "https://example.com/train.py" hf jobs run uv "https://example.com/train.py" --flavor a10g-large hf jobs uv run "https://example.com/train.py" --flavor a10g-large hf jobs uv run --secret HF_TOKEN "https://example.com/train.py" hf jobs uv run \ hf jobs ps # List all jobs
What does the hugging-face-model-trainer skill do?
Train or fine-tune TRL language models on Hugging Face Jobs, including SFT, DPO, GRPO, and GGUF export.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill hugging-face-model-trainer-sickn33-antigravity-awesome --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.