Quickstart

Fine-tuning builds upon the foundational knowledge of pre-trained models, enabling them to specialize in specific tasks tailored to your requirements. With Together, you can fine-tune and run your model in the cloud without having to set up any GPUs. The Together API allows you to enhance state-of-the-art models sourced from leading open-source platforms using your unique data. You can tune all model layers, control hyper-parameters, and run the model with Together’s inference service. You can also download the weights to run wherever you’d like.

We offer two ways to fine-tune a model:

• CLI
• Python library

Choose whichever best fits your needs!