hugging-face-jobs
Run compute jobs and Python scripts on HF infrastructure
Developer Setup
Setup & Installation
npx skills add https://github.com/huggingface/skills --skill hugging-face-jobsnpx skills add https://github.com/huggingface/skills --skill hugging-face-jobsOverview
What This Skill Does
Runs Python workloads on Hugging Face managed infrastructure — CPUs, GPUs, or TPUs — without any local setup. Supports UV scripts with inline dependencies, Docker-based jobs, scheduled tasks, and result persistence to the Hugging Face Hub. Handles authentication, secrets management, hardware selection, and timeout configuration.
Application
When to use this Skill
- Configuring integration settings for custom agent workflows.
- Optimizing query execution and response latency in production.
- Developing clean, standard-compliant implementations for enterprise services.
- Troubleshooting connection timeouts and authentication handshakes.
- Monitoring API rate limits and execution pipelines programmatically.
Documentation
Show Skills.md file
Hugging Face ZeroGPU
Rules and patterns for ML demos on Hugging Face Spaces with ZeroGPU hardware. Covers @spaces.GPU, duration and quota tuning, process isolation, the CUDA availability model, concurrency safety, and CUDA build constraints.
Scope
This skill is for Gradio SDK Spaces using ZeroGPU hardware. Docker and Static Spaces cannot schedule onto ZeroGPU, and Streamlit apps now run as Docker Spaces — so this skill applies only to Gradio. For general Gradio coding (components, layouts, event listeners), see the huggingface-gradio skill in this repo. The authoritative ZeroGPU docs live at https://huggingface.co/docs/hub/spaces-zerogpu — refer to them for the current backing GPU, runtime version lists, and tier thresholds, all of which change over time.
Reference Files
| Reference | When to read |
|---|---|
references/concurrency.md |
Always read alongside SKILL.md when writing ZeroGPU code — handlers run in parallel by default |
references/how-zerogpu-works.md |
When reasoning about cold-starts, worker reuse, why module-scope warmup does not carry to requests, or why returning CUDA tensors hangs |
references/how-quota-works.md |
When choosing duration values, debugging illegal duration vs quota exceeded errors, or explaining why default 60s blocks short tasks |
references/cuda-and-deps.md |
When installing CUDA-dependent packages (e.g. flash-attn), pinning torch side-cars, or reading wheel filename tags |
Hardware
ZeroGPU exposes two GPU sizes that map to a fraction of the backing card:
size |
Slice of backing GPU | Quota cost |
|---|---|---|
large (default) |
Half | 1x |
xlarge |
Full | 2x |
Recommendations