transformers - Work with pre-trained transformer models and fine-tune.
Access pre-trained transformer models for NLP, vision, audio, and multimodal tasks, including inference, tokenization, and fine-tuning.
Tags
Updated: 2026-09-16Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Load pre-trained models
- Run pipeline inference
- Generate text using decoding strategies
- Fine-tune models on custom data
- Tokenize text for model inputs
Inputs
- Pre-trained model identifiers
- Input text, images, or audio
- Custom training datasets
- Hugging Face access tokens
Outputs
- Model inference outputs
- Generated text
- Fine-tuned model checkpoints
- Tokenized input tensors
Requirements
- Python environment with torch and transformers
- Hugging Face access token for gated models