Skip to content

Fine-tuning

Fine-tuning will let you adapt a model from the catalog to your own data, so it follows your formats, tone and domain vocabulary without having to be told in every prompt.

  1. Prepare a dataset of example exchanges and upload it via Datasets.
  2. Pick a base model from the models marked as fine-tunable.
  3. Start a job, choosing how long to train and how strongly to adapt.
  4. Evaluate the result against a held-out split before promoting it.
  5. Call it by its own model ID, using the same API and keys as any other model.
  • Adapters, not full retrains — cheaper and faster than training from scratch.
  • Your tuned model is private to your workspace.
  • Training is billed separately from inference, since it consumes GPU time rather than tokens.
  • Your training data belongs to you, and is not used to train anything else.

Consistent structured output, a house style or tone, domain jargon, or classification where prompting alone is inconsistent.

Fine-tuning is a poor fit for teaching a model new facts — for that, retrieval over embeddings is usually the better tool.