Create a fine-tuning job
Creates a job that fine-tunes a specified model based on a given dataset.
Authorizations
Bearer authentication header of the form Bearer <token>, where <token> is your auth token.
Query Parameters
Project ID to list fine tuning jobs
Body
The name of the model to fine-tune
"meta-llama/Llama-3.1-70B-Instruct"
"meta-llama/Llama-3.1-70B"
"meta-llama/Llama-3.1-8B-Instruct"
"meta-llama/Llama-3.1-8B"
The ID of the uploaded training data file
"file_123456"
Checkpoint to continue training from
- HFCheckpointParametersRequest
- FTCheckpointParameters
The ID of the uploaded validation data file
"file_123456"
Training hyperparameters
Fine-tuning method configuration
- SupervisedMethodConfig
- SpecDraftMethodConfig
Job integration paramaters
- WandbIntegrationRequest
- MlflowIntegrationRequest
- HfExportIntegrationRequest
The seed controls the reproducibility of the job
42
The suffix that will be used for output model name
64List of tags to associate with the job
"my-ft-job"
Extra params to pass to the fine-tuning job
Response
Successful Response
validating_files, queued, running, succeeded, failed, cancelled - SupervisedMethodConfig
- SpecDraftMethodConfig
- WandbIntegrationResponse
- MlflowIntegrationResponse
- HfExportIntegrationResponse
- HFCheckpointParametersResponse
- FTCheckpointParameters
fine_tuning.job "fine_tuning.job"64