hugging-face-jobs

Submit and manage computational workloads on Hugging Face Jobs infrastructure.

Updated Mar 5, 2026
One-click install
npx skills add https://github.com/Himanshu040604/codex-skills-setup --skill hugging-face-jobs-himanshu040604
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hugging-face-jobs
Source: https://github.com/Himanshu040604/codex-skills-setup/tree/main/assets/codex/skills/claude-import/skills/plugins/huggingface-skills%40claude-plugins-official/skills/hugging-face-jobs
Command: npx skills add https://github.com/Himanshu040604/codex-skills-setup --skill hugging-face-jobs-himanshu040604

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill allows users to run computational workloads, including data processing, model inference, and experiments, on Hugging Face's managed infrastructure without needing local setup.

Core Features & Use Cases

  • Cloud Compute: Execute Python scripts or Docker containers on cloud CPUs and GPUs.
  • Scalability: Choose from various hardware flavors (CPU, GPU, TPU) based on workload needs.
  • Persistence: Ensure results are saved by pushing them to the Hugging Face Hub.
  • Authentication: Securely handle Hugging Face tokens for Hub interactions.
  • Use Case: Run a batch inference job on a large dataset using a GPU, then automatically push the generated predictions as a new dataset to the Hugging Face Hub.

Quick Start

Use the hugging-face-jobs skill to run a Python script named 'my_script.py' on a basic CPU instance with a 30-minute timeout.

Frequently Asked Questions about hugging-face-jobs

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

FAQPage Schema
How do I run Python scripts on cloud GPU instances using Hugging Face Jobs?

You can run Python scripts on cloud GPU instances by submitting UV scripts to Hugging Face Jobs with your specified hardware flavor, authentication token, and timeout configuration to execute workloads without local setup.

Can I execute custom Docker containers on Hugging Face compute infrastructure?

Yes, you can execute custom Docker containers on Hugging Face compute infrastructure by submitting them as computational workloads, selecting appropriate CPU or GPU hardware flavors, and managing task timeouts.

Does Hugging Face Jobs support automatically pushing batch inference results to the Hub?

Hugging Face Jobs supports automatically pushing batch inference results to the Hub by securely handling authentication via HF_TOKEN secrets and persisting generated outputs directly to your repository.

What hardware flavors can I choose when running workloads on Hugging Face Jobs?

When running workloads on Hugging Face Jobs, you can choose from various cloud compute hardware flavors including CPU, GPU, and TPU instances based on your specific data processing or model inference requirements.

Do I need to set up local GPUs to run model inference jobs with Hugging Face?

You do not need to set up local GPUs to run model inference jobs because Hugging Face Jobs provides managed cloud compute infrastructure, allowing deterministic task execution entirely through remote hardware selection.

How do I configure authentication for cloud compute tasks on Hugging Face infrastructure?

You configure authentication for cloud compute tasks on Hugging Face infrastructure by managing HF_TOKEN secrets, which securely handle Hub interactions and ensure your results are properly persisted after execution.