skills.rest
allanninal
allanninalCOMMUNITY
Explore all skills by this author →
Website𝕏@allanninGitHub

huggingface-skills

Manage Hugging Face Hub models, datasets, and Spaces via CLI or Python API.

1|Updated Jan 31, 2026
Search Tags:#machine learning#model training#nlp#datasets#huggingface#gguf#trl
One-click install
npx skills add https://github.com/allanninal/claude-code-skills --skill huggingface-skills
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: huggingface-skills
Source: https://github.com/allanninal/claude-code-skills/tree/main/skills/huggingface-skills
Command: npx skills add https://github.com/allanninal/claude-code-skills --skill huggingface-skills

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill streamlines operations on the Hugging Face Hub, simplifying the management of models, datasets, and Spaces, as well as facilitating ML training and evaluation.

Core Features & Use Cases

  • Model & Dataset Management: Upload, download, and create repositories for models and datasets.
  • ML Training: Train models using TRL (SFT, DPO, GRPO).
  • Model Evaluation: Run evaluations using the lighteval framework.
  • Space Management: Create and manage Gradio or Streamlit Spaces.
  • GGUF Conversion: Convert Hugging Face models to GGUF format for local inference.

Quick Start

Use the huggingface-skills to upload the local directory 'my-model-files' to a new model repository named 'my-username/my-new-model' on Hugging Face Hub.

Frequently Asked Questions about huggingface-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I train a Hugging Face model using TRL?▼

To train a Hugging Face model using TRL, this Skill facilitates Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and GRPO workflows. It integrates the trl library to manage and execute machine learning training operations directly.

Can I convert a Hugging Face model to GGUF format for local inference?▼

Yes, you can convert Hugging Face models to GGUF format for local inference. This Skill supports GGUF conversion operations, allowing you to prepare and optimize transformer models for deployment in local inference environments.

What's the best way to manage Hugging Face Hub repositories?▼

The best way to manage Hugging Face Hub repositories is through CLI and Python API interactions. This Skill streamlines uploading, downloading, and creating repositories for both models and datasets directly on the Hub.

Does this workflow support running model evaluations with lighteval?▼

Yes, this workflow supports running model evaluations using the lighteval framework. It integrates the library to facilitate comprehensive evaluation of your machine learning models within the Hugging Face ecosystem.

How do I create and manage Hugging Face Spaces for Gradio or Streamlit?▼

To create and manage Hugging Face Spaces, this Skill supports deployment for both Gradio and Streamlit applications. It handles Space creation and management operations through CLI and Python API interactions.

What libraries do I need to interact with Hugging Face datasets and transformers?▼

You need to integrate with the transformers, datasets, trl, and lighteval libraries. This Skill leverages these frameworks to support comprehensive machine learning workflows, including data loading, model training, and evaluation.

Related Skills

Explore More Software Engineering Skills→
Brilly-BohyunBrilly-Bohyun

huggingface-hub

Manage models, datasets, and repositories on the Hugging Face Hub.

Community
Advanced
Andrew-GirgisAndrew-Girgis

huggingface-spaces

Build, deploy, and maintain Hugging Face Spaces with Gradio, Docker, or Static SDKs.

Community
Advanced
ovachieverovachiever
52

transformers

Load, infer, and fine-tune pre-trained transformer models across modalities.

Community
Advanced
luokai25luokai25

hf-cli

Manage Hugging Face Hub repositories, models, datasets, and Spaces.

Community
Intermediate