This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on 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, and model persistence. Should be invoked for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.
mog install mogteam/hf-model-trainer
Sourced from huggingface/skills under the Apache 2.0 license.
After installing, your AI assistant needs a pointer to the skill file. Use the agent card below or the --wire flag.
Install + auto-wire
mog install mogteam/hf-model-trainer --target cursor --wireCreates .cursor/rules/hf-model-trainer.mdc pointing to the skill.
Agent card
## Hf Model Trainer This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on 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, and model persistence. Should be invoked for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup. - **Install**: `mog install mogteam/hf-model-trainer --target cursor` - **Type**: skill - **Path after install**: `.cursor/skills/hf-model-trainer/SKILL.md` - **Targets**: cursor, claude-code, codex, gemini-cli, windsurf, generic When editing, read and follow @.cursor/skills/hf-model-trainer/SKILL.md
Paste into AGENTS.md, .cursor/rules, or your agent's instructions.
Install paths by target
No ratings yet.

This skill should be used when users want to train or fine-tune language models using TRL (Transformer Reinforcement Learning) on 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, and model persistence. Should be invoked for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.
Free
mog install mogteam/hf-model-trainerSource
Hugging Face
github.com/huggingface/skillsThis package is sourced from the above repository and distributed on mog under its original license.
@mogteam
Official curated skills and prompt templates from the mog.md team.