Load models, run inference, and fine-tune with Transformers.
Develop model-loading, inference, text-generation, and fine-tuning code with Hugging Face Transformers.
When to use it
Use when working with Transformers model classes, pipelines, tokenizers, generation configurations, or training arguments.
It changes how the agent implements model loading, inference, generation, tokenization, and fine-tuning in a project.
What you provide
Requires Python 3.10+; the installation procedure selects Python 3.11.
Requires transformers 5.18.0, installed with its torch extra.
This skill
Hugging Face Hub
Downloads models from Hugging Face Hub
arXiv
Fetches the author's paper record