Fine-tuning Jobs

Create a new fine-tuning job

Changed on
post/v1/fine-tuning

Request

Request body

base_model_namestring required
dataset_namestring required
namestring required

The name of the fine-tuning job

type'LORA' | 'QLORA' required

The type of fine-tuning job to run

Example request

{
  "parameters": {
    "batch_size": 8,
    "num_epochs": 3,
    "shuffle": true,
    "lr": 0.0003,
    "seed": 42
  }
}

Response

Created

idstring uuid
created_atstring date-time
updated_atstring date-time
status'NEW' | 'QUEUED' | 'RUNNING' | 'COMPLETED' | 'FAILED' | 'STOPPING' | 'STOPPED'
base_model_namestring
dataset_namestring
namestring
type'LORA' | 'QLORA'
current_stepinteger
total_stepsinteger
current_epochinteger
total_epochsinteger
num_tokensinteger

Number of tokens processed in the fine-tuning job

Changes

    • breaking

      removed the enum value FULL of the request property

    • info

      removed the FULL enum value from the response property for the response status

    • breaking

      added the new required request property

    • warning

      removed the request property /

    • warning

      added the new QUEUED enum value to the response property for the response status

    • info

      added the new optional request property /

    • info

      added the new optional request property /

    • info

      added the new optional request property /

    • info

      added the new optional request property /

    • info

      added the optional property to the response with the status

    • info

      removed the PENDING enum value from the response property for the response status