Create fine-tuning job
Creates a fine-tuning job which begins the process of creating a new model from a given dataset. The response includes details of the enqueued job, including job status and the name of the fine-tuned model once complete.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Body
The name of the model to fine-tune. Options: 'llama-3.1-8b' or 'llama-3.1-70b'.
The name of the base model whose output will be styled. Options: 'gpt-4o', 'gpt-4o-2024-05-13', 'gpt-4o-2024-08-06', 'claude-3-5-sonnet-20240620', 'claude-3-opus-20240229', 'other'. Use 'other' for any other model.
Array of IDs of uploaded files to use for training. All files must be text-only.
A string of up to 16 characters that will be added to your fine-tuned model name. For example, a suffix of 'custom-model-name' would produce a model name like 'ftm:llama-3.1-70b:custom-model-name:0b5ace7e-ffb2-43cf-8a4d-0af17bc2b1d1'.
The number of training examples to use for training. Can be an integer or 'auto'. Higher numbers increase quality, but also increase cost.
If true, the model will be trained to maintain consistency of length, where the styled output is approximately the same length as the input.
Response
The type of the object, always 'fine_tuning.job'.
The ID of the fine-tuning job.
The model being fine-tuned.
The name of the base model whose output is to be styled.
The Unix timestamp for when the job was created.
The name of the fine-tuned model.
The organization ID associated with this fine-tuning job.
The current status of the job.
Array of file IDs used for training.