huggingface-vision-trainer
huggingfaceTrain vision models without local GPU setup.
vision model training
What it does
Fine-tunes vision models on managed Hugging Face Jobs GPUs and saves the results to the Hugging Face Hub.
When to use it
Use this skill for vision-model training or fine-tuning involving detection, classification, SAM/SAM2 segmentation, image matting, or Hugging Face Jobs.
How to use it
Give it a vision training request and configuration; it prepares a training script, submits a Hugging Face Jobs run, and returns job and monitoring details.
What you provide
- vision task
- Hub dataset ID
- model name
- Hub model ID
Uses
Access · 4
This skill
Hugging Face Hub
Pushes models to the Hub
Hugging Face Hub datasets
Validates Hub datasets
Hugging Face Jobs
Starts cloud GPU jobs
uv
Helper scripts with uv
What you need · 2
A Hugging Face Pro, Team, or Enterprise account is required because Jobs requires a paid plan.
The dataset to train on must already exist on the Hugging Face Hub.
About this skill
- Visibility
- Public
- Repository
- huggingface/skills
- Created
- Oct 8, 2026
- Updated
- Oct 8, 2026
- Files
- 12