Cog's training API allows you to define a fine-tuning interface for an existing Cog model, so users of the model can bring their own training data to create derivative fune-tuned models. Real-world examples of this API in use include fine-tuning SDXL with images or fine-tuning Llama 2 with structured text.
See the Cog training reference docs for more details.
This simple trainable model takes a string as input and returns a string as output.
Then you can run it like this:
cog train -i prefix=helloCheck out these guides to learn how to fine-tune models on Replicate: