Data
Format adapter
What is format adapter?
A format adapter converts a dataset from the shape you have into the shape the trainer requires — for example conversation records into a model’s native chat format, or preference records into training pairs. It is a pure transformation with no learning involved.
Adapters exist so the dataset you upload does not have to be the dataset the trainer consumes. That separation means changing the target model does not require rebuilding your data.
A good adapter fails loudly. Silently dropping malformed records produces a smaller dataset than you think you have and a model trained on an unrepresentative subset of it.
Related terms
Start with the free tier
A magic link creates your account, your tenant and your first API key. No card until you ask for compute.