Path parameters

• fine_tuning_job_id string Required

  The ID of the fine-tuning job.

Responses

• 200 application/json

  OK

  Hide response attributes Show response attributes object

  • id string Required

    The object identifier, which can be referenced in the API endpoints.

  • created_at integer Required

    The Unix timestamp (in seconds) for when the fine-tuning job was created.

  • error object | null Required

    For fine-tuning jobs that have failed, this will contain more information on the cause of the failure.

    Hide error attributes Show error attributes object | null

    • code string Required

      A machine-readable error code.

    • message string Required

      A human-readable error message.

    • param string | null Required

      The parameter that was invalid, usually training_file or validation_file. This field will be null if the failure was not parameter-specific.

  • fine_tuned_model string | null Required

    The name of the fine-tuned model that is being created. The value will be null if the fine-tuning job is still running.

  • finished_at integer | null Required

    The Unix timestamp (in seconds) for when the fine-tuning job was finished. The value will be null if the fine-tuning job is still running.

  • hyperparameters object Required

    The hyperparameters used for the fine-tuning job. This value will only be returned when running supervised jobs.

    Hide hyperparameters attributes Show hyperparameters attributes object

    • batch_size string | integer

      Number of examples in each batch. A larger batch size means that model parameters are updated less frequently, but with lower variance.

      One of:

      string-1 string integer-2 integer

      Value is auto. Default value is auto.

      Minimum value is 1, maximum value is 256. Default value is auto.

    • learning_rate_multiplier string | number

      Scaling factor for the learning rate. A smaller learning rate may be useful to avoid overfitting.

      One of:

      string-1 string number-2 number

      Value is auto. Default value is auto.

      Minimum value is 0. Default value is auto.

    • n_epochs string | integer

      The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.

      One of:

      string-1 string integer-2 integer

      Value is auto. Default value is auto.

      Minimum value is 1, maximum value is 50. Default value is auto.

  • model string Required

    The base model that is being fine-tuned.

  • object string Required

    The object type, which is always "fine_tuning.job".

    Value is fine_tuning.job.

  • organization_id string Required

    The organization that owns the fine-tuning job.

  • result_files array[string] Required

    The compiled results file ID(s) for the fine-tuning job. You can retrieve the results with the Files API.

  • status string Required

    The current status of the fine-tuning job, which can be either validating_files, queued, running, succeeded, failed, or cancelled.

    Values are validating_files, queued, running, succeeded, failed, or cancelled.

  • trained_tokens integer | null Required

    The total number of billable tokens processed by this fine-tuning job. The value will be null if the fine-tuning job is still running.

  • training_file string Required

    The file ID used for training. You can retrieve the training data with the Files API.

  • validation_file string | null Required

    The file ID used for validation. You can retrieve the validation results with the Files API.

  • integrations array[object] | null

    A list of integrations to enable for this fine-tuning job.

    Not more than 5 elements.

    Hide integrations attributes Show integrations attributes object

    • type string Required

      The type of the integration being enabled for the fine-tuning job

      Value is wandb.

    • wandb object Required

      The settings for your integration with Weights and Biases. This payload specifies the project that metrics will be sent to. Optionally, you can set an explicit display name for your run, add tags to your run, and set a default entity (team, username, etc) to be associated with your run.

      Hide wandb attributes Show wandb attributes object

      • project string Required

        The name of the project that the new run will be created under.

      • name string | null

        A display name to set for the run. If not set, we will use the Job ID as the name.

      • entity string | null

        The entity to use for the run. This allows you to set the team or username of the WandB user that you would like associated with the run. If not set, the default entity for the registered WandB API key is used.

      • tags array[string]

        A list of tags to be attached to the newly created run. These tags are passed through directly to WandB. Some default tags are generated by OpenAI: "openai/finetune", "openai/{base-model}", "openai/{ftjob-abcdef}".

  • seed integer Required

    The seed used for the fine-tuning job.

  • estimated_finish integer | null

    The Unix timestamp (in seconds) for when the fine-tuning job is estimated to finish. The value will be null if the fine-tuning job is not running.

  • method object

    The method used for fine-tuning.

    Hide method attributes Show method attributes object

    • type string

      The type of method. Is either supervised or dpo.

      Values are supervised or dpo.

    • supervised object

      Configuration for the supervised fine-tuning method.

      Hide supervised attribute Show supervised attribute object

      • hyperparameters object

        The hyperparameters used for the fine-tuning job.

        Hide hyperparameters attributes Show hyperparameters attributes object

        • batch_size string | integer

          Number of examples in each batch. A larger batch size means that model parameters are updated less frequently, but with lower variance.

          One of:

          string-1 string integer-2 integer

          Value is auto. Default value is auto.

          Minimum value is 1, maximum value is 256. Default value is auto.

        • learning_rate_multiplier string | number

          Scaling factor for the learning rate. A smaller learning rate may be useful to avoid overfitting.

          One of:

          string-1 string number-2 number

          Value is auto. Default value is auto.

          Minimum value is 0. Default value is auto.

        • n_epochs string | integer

          The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.

          One of:

          string-1 string integer-2 integer

          Value is auto. Default value is auto.

          Minimum value is 1, maximum value is 50. Default value is auto.

    • dpo object

      Configuration for the DPO fine-tuning method.

      Hide dpo attribute Show dpo attribute object

      • hyperparameters object

        The hyperparameters used for the fine-tuning job.

        Hide hyperparameters attributes Show hyperparameters attributes object

        • beta string | number

          The beta value for the DPO method. A higher beta value will increase the weight of the penalty between the policy and reference model.

          One of:

          string-1 string number-2 number

          Value is auto. Default value is auto.

          Minimum value is 0, maximum value is 2. Default value is auto.

        • batch_size string | integer

          Number of examples in each batch. A larger batch size means that model parameters are updated less frequently, but with lower variance.

          One of:

          string-1 string integer-2 integer

          Value is auto. Default value is auto.

          Minimum value is 1, maximum value is 256. Default value is auto.

        • learning_rate_multiplier string | number

          Scaling factor for the learning rate. A smaller learning rate may be useful to avoid overfitting.

          One of:

          string-1 string number-2 number

          Value is auto. Default value is auto.

          Minimum value is 0. Default value is auto.

        • n_epochs string | integer

          The number of epochs to train the model for. An epoch refers to one full cycle through the training dataset.

          One of:

          string-1 string integer-2 integer

          Value is auto. Default value is auto.

          Minimum value is 1, maximum value is 50. Default value is auto.

  • metadata object | null

    Set of 16 key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format, and querying for objects via API or the dashboard.

    Keys are strings with a maximum length of 64 characters. Values are strings with a maximum length of 512 characters.

    Hide metadata attribute Show metadata attribute object | null

    • * string Additional properties