hugging-face-cli

Manage Hugging Face Hub repositories, downloads, uploads, and cloud compute jobs via the hf CLI.

1|Updated Jan 22, 2026

Search Tags:#download #repositories #upload #huggingface #hub #hf-cli #compute-jobs

Ready to run

npx skills add https://github.com/sitammeur/ml-agent-skills --skill hugging-face-cli-sitammeur

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: hugging-face-cli
Source: https://github.com/sitammeur/ml-agent-skills/tree/main/skills/hugging-face-cli
Command: npx skills add https://github.com/sitammeur/ml-agent-skills --skill hugging-face-cli-sitammeur

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables developers and data teams to perform Hugging Face Hub operations directly from the command line by leveraging the hf CLI, reducing manual context switching and API complexity.

Core Features & Use Cases

Hub management: Download models, datasets, and spaces; upload files; and manage repositories from your terminal.
Compute and deployment: Run cloud compute jobs and manage endpoints via CLI commands.
Automation & scripts: Integrate with CI/CD to authenticate, list resources, and automate common Hub workflows.

Use Case: When you need to batch-download a set of models for an offline evaluation, or publish a new dataset to HF Hub from a local workspace.

Quick Start

Use the hf CLI to login, download a model to local cache, and list your repos.

Frequently Asked Questions about hugging-face-cli

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

FAQPage Schema

How do I download models and datasets from the Hugging Face Hub using the command line?▼

To download Hugging Face Hub resources, use the hf CLI to fetch models, datasets, and spaces directly to your local cache. This approach reduces manual context switching by automating terminal-based retrieval for offline workflows.

Can I run cloud compute jobs on Hugging Face directly from my terminal?▼

Yes, you can run cloud compute jobs on Hugging Face using specific hf CLI commands. The CLI allows you to manage compute tasks and endpoints, enabling deployment and execution without leaving your terminal environment.

What is the best way to automate uploading files to a Hugging Face repository?▼

Automating Hugging Face uploads is best achieved by integrating the hf CLI into your CI/CD pipelines. You can authenticate, create repositories, and publish files or datasets from your local workspace automatically.

Do I need API tokens to manage Hugging Face Hub repositories via CLI?▼

Yes, authentication is required to manage Hugging Face Hub repositories. You must use the hf CLI to login with your tokens before listing resources, uploading files, or running cloud compute jobs in automated workflows.

How does managing local cache work when batch-downloading Hugging Face models?▼

Managing local cache with the hf CLI involves terminal commands that store downloaded Hugging Face models and datasets efficiently. This prevents redundant downloads and streamlines offline evaluations across multiple resources.