openai.resources.fine_tuning

The fine_tuning module provides classes for handling fine-tuning operations, including the initiation, management, and retrieval of fine-tuning jobs.

The module supports synchronous and asynchronous operations, offering interfaces for working with jobs directly, as well as with raw or streaming responses. Designed for use in applications requiring custom model training on specific datasets to improve model performance for tailored tasks.

Modules:

Name	Description
fine_tuning	The fine_tuning module provides classes for handling fine-tuning operations, including the initiation, management, and retrieval of fine-tuning jobs. The module supports synchronous and asynchronous operations, offering interfaces for working with jobs directly, as well as with raw or streaming responses. Designed for use in applications requiring custom model training on specific datasets to improve model performance for tailored tasks.
jobs	The jobs module provides synchronous and asynchronous access to fine-tuning job resources and enables you to create, retrieve, list, and cancel fine-tuning jobs and list the events associated with them. Fine-tuning jobs allow you to customize pre-trained models with your own datasets, optimizing performance for specific tasks or improving understanding of particular data types. The classes in the jobs module includes methods for managing fine-tuning jobs, like creating a new job, fetching details of an existing job, listing all jobs, canceling a job, and listing events associated with a job.

Classes:

Name	Description
AsyncFineTuning
AsyncFineTuningWithRawResponse
AsyncFineTuningWithStreamingResponse
AsyncJobs
AsyncJobsWithRawResponse
AsyncJobsWithStreamingResponse
FineTuning
FineTuningWithRawResponse
FineTuningWithStreamingResponse
Jobs
JobsWithRawResponse
JobsWithStreamingResponse

AsyncFineTuning

AsyncFineTuning(client: AsyncOpenAI)

Methods:

Name	Description
jobs
with_raw_response
with_streaming_response

jobs

jobs() -> AsyncJobs

with_raw_response

with_raw_response() -> AsyncFineTuningWithRawResponse

with_streaming_response

with_streaming_response() -> (
    AsyncFineTuningWithStreamingResponse
)

AsyncFineTuningWithRawResponse

AsyncFineTuningWithRawResponse(
    fine_tuning: AsyncFineTuning,
)

Methods:

Name	Description
jobs

jobs

jobs() -> AsyncJobsWithRawResponse

AsyncFineTuningWithStreamingResponse

AsyncFineTuningWithStreamingResponse(
    fine_tuning: AsyncFineTuning,
)

Methods:

Name	Description
jobs

jobs

jobs() -> AsyncJobsWithStreamingResponse

AsyncJobs

AsyncJobs(client: AsyncOpenAI)

Methods:

Name	Description
cancel	Immediately cancel a fine-tune job.
create	Creates a fine-tuning job which begins the process of creating a new model from
list	List your organization's fine-tuning jobs
list_events	Get status updates for a fine-tuning job.
retrieve	Get info about a fine-tuning job.
with_raw_response
with_streaming_response

cancel async

cancel(
    fine_tuning_job_id: str,
    *,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> FineTuningJob

Immediately cancel a fine-tune job.

Parameters:

Name	Type	Description	Default
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

create async

create(
    *,
    model: Union[
        str,
        Literal[
            "babbage-002", "davinci-002", "gpt-3.5-turbo"
        ],
    ],
    training_file: str,
    hyperparameters: Hyperparameters | NotGiven = NOT_GIVEN,
    suffix: Optional[str] | NotGiven = NOT_GIVEN,
    validation_file: Optional[str] | NotGiven = NOT_GIVEN,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> FineTuningJob

Creates a fine-tuning job which begins the process of creating a new model from a given dataset.

Response includes details of the enqueued job including job status and the name of the fine-tuned models once complete.

Learn more about fine-tuning

Parameters:

Name	Type	Description	Default
model	Union[str, Literal['babbage-002', 'davinci-002', 'gpt-3.5-turbo']]	The name of the model to fine-tune. You can select one of the supported models.	required
training_file	str	The ID of an uploaded file that contains training data. See upload file for how to upload a file. Your dataset must be formatted as a JSONL file. Additionally, you must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.	required
hyperparameters	Hyperparameters | NotGiven	The hyperparameters used for the fine-tuning job.	NOT_GIVEN
suffix	Optional[str] | NotGiven	A string of up to 18 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 ft:gpt-3.5-turbo:openai:custom-model-name:7p4lURel.	NOT_GIVEN
validation_file	Optional[str] | NotGiven	The ID of an uploaded file that contains validation data. If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. The same data should not be present in both train and validation files. Your dataset must be formatted as a JSONL file. You must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.	NOT_GIVEN
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

list

list(
    *,
    after: str | NotGiven = NOT_GIVEN,
    limit: int | NotGiven = NOT_GIVEN,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> AsyncPaginator[
    FineTuningJob, AsyncCursorPage[FineTuningJob]
]

List your organization's fine-tuning jobs

Parameters:

Name	Type	Description	Default
after	str | NotGiven	Identifier for the last job from the previous pagination request.	NOT_GIVEN
limit	int | NotGiven	Number of fine-tuning jobs to retrieve.	NOT_GIVEN
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

list_events

list_events(
    fine_tuning_job_id: str,
    *,
    after: str | NotGiven = NOT_GIVEN,
    limit: int | NotGiven = NOT_GIVEN,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> AsyncPaginator[
    FineTuningJobEvent, AsyncCursorPage[FineTuningJobEvent]
]

Get status updates for a fine-tuning job.

Parameters:

Name	Type	Description	Default
after	str | NotGiven	Identifier for the last event from the previous pagination request.	NOT_GIVEN
limit	int | NotGiven	Number of events to retrieve.	NOT_GIVEN
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

retrieve async

retrieve(
    fine_tuning_job_id: str,
    *,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> FineTuningJob

Get info about a fine-tuning job.

Learn more about fine-tuning

Parameters:

Name	Type	Description	Default
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

with_raw_response

with_raw_response() -> AsyncJobsWithRawResponse

with_streaming_response

with_streaming_response() -> AsyncJobsWithStreamingResponse

AsyncJobsWithRawResponse

AsyncJobsWithRawResponse(jobs: AsyncJobs)

Attributes:

Name	Type	Description
cancel
create
list
list_events
retrieve

cancel instance-attribute

cancel = async_to_raw_response_wrapper(cancel)

create instance-attribute

create = async_to_raw_response_wrapper(create)

list instance-attribute

list = async_to_raw_response_wrapper(list)

list_events instance-attribute

list_events = async_to_raw_response_wrapper(list_events)

retrieve instance-attribute

retrieve = async_to_raw_response_wrapper(retrieve)

AsyncJobsWithStreamingResponse

AsyncJobsWithStreamingResponse(jobs: AsyncJobs)

Attributes:

Name	Type	Description
cancel
create
list
list_events
retrieve

cancel instance-attribute

cancel = async_to_streamed_response_wrapper(cancel)

create instance-attribute

create = async_to_streamed_response_wrapper(create)

list instance-attribute

list = async_to_streamed_response_wrapper(list)

list_events instance-attribute

list_events = async_to_streamed_response_wrapper(
    list_events
)

retrieve instance-attribute

retrieve = async_to_streamed_response_wrapper(retrieve)

FineTuning

FineTuning(client: OpenAI)

Methods:

Name	Description
jobs
with_raw_response
with_streaming_response

jobs

jobs() -> Jobs

with_raw_response

with_raw_response() -> FineTuningWithRawResponse

with_streaming_response

with_streaming_response() -> (
    FineTuningWithStreamingResponse
)

FineTuningWithRawResponse

FineTuningWithRawResponse(fine_tuning: FineTuning)

Methods:

Name	Description
jobs

jobs

jobs() -> JobsWithRawResponse

FineTuningWithStreamingResponse

FineTuningWithStreamingResponse(fine_tuning: FineTuning)

Methods:

Name	Description
jobs

jobs

jobs() -> JobsWithStreamingResponse

Jobs

Jobs(client: OpenAI)

Methods:

Name	Description
cancel	Immediately cancel a fine-tune job.
create	Creates a fine-tuning job which begins the process of creating a new model from
list	List your organization's fine-tuning jobs
list_events	Get status updates for a fine-tuning job.
retrieve	Get info about a fine-tuning job.
with_raw_response
with_streaming_response

cancel

cancel(
    fine_tuning_job_id: str,
    *,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> FineTuningJob

Immediately cancel a fine-tune job.

Parameters:

Name	Type	Description	Default
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

create

create(
    *,
    model: Union[
        str,
        Literal[
            "babbage-002", "davinci-002", "gpt-3.5-turbo"
        ],
    ],
    training_file: str,
    hyperparameters: Hyperparameters | NotGiven = NOT_GIVEN,
    suffix: Optional[str] | NotGiven = NOT_GIVEN,
    validation_file: Optional[str] | NotGiven = NOT_GIVEN,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> FineTuningJob

Creates a fine-tuning job which begins the process of creating a new model from a given dataset.

Response includes details of the enqueued job including job status and the name of the fine-tuned models once complete.

Learn more about fine-tuning

Parameters:

Name	Type	Description	Default
model	Union[str, Literal['babbage-002', 'davinci-002', 'gpt-3.5-turbo']]	The name of the model to fine-tune. You can select one of the supported models.	required
training_file	str	The ID of an uploaded file that contains training data. See upload file for how to upload a file. Your dataset must be formatted as a JSONL file. Additionally, you must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.	required
hyperparameters	Hyperparameters | NotGiven	The hyperparameters used for the fine-tuning job.	NOT_GIVEN
suffix	Optional[str] | NotGiven	A string of up to 18 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 ft:gpt-3.5-turbo:openai:custom-model-name:7p4lURel.	NOT_GIVEN
validation_file	Optional[str] | NotGiven	The ID of an uploaded file that contains validation data. If you provide this file, the data is used to generate validation metrics periodically during fine-tuning. These metrics can be viewed in the fine-tuning results file. The same data should not be present in both train and validation files. Your dataset must be formatted as a JSONL file. You must upload your file with the purpose fine-tune. See the fine-tuning guide for more details.	NOT_GIVEN
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

list

list(
    *,
    after: str | NotGiven = NOT_GIVEN,
    limit: int | NotGiven = NOT_GIVEN,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> SyncCursorPage[FineTuningJob]

List your organization's fine-tuning jobs

Parameters:

Name	Type	Description	Default
after	str | NotGiven	Identifier for the last job from the previous pagination request.	NOT_GIVEN
limit	int | NotGiven	Number of fine-tuning jobs to retrieve.	NOT_GIVEN
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

list_events

list_events(
    fine_tuning_job_id: str,
    *,
    after: str | NotGiven = NOT_GIVEN,
    limit: int | NotGiven = NOT_GIVEN,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> SyncCursorPage[FineTuningJobEvent]

Get status updates for a fine-tuning job.

Parameters:

Name	Type	Description	Default
after	str | NotGiven	Identifier for the last event from the previous pagination request.	NOT_GIVEN
limit	int | NotGiven	Number of events to retrieve.	NOT_GIVEN
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

retrieve

retrieve(
    fine_tuning_job_id: str,
    *,
    extra_headers: Headers | None = None,
    extra_query: Query | None = None,
    extra_body: Body | None = None,
    timeout: float | Timeout | None | NotGiven = NOT_GIVEN
) -> FineTuningJob

Get info about a fine-tuning job.

Learn more about fine-tuning

Parameters:

Name	Type	Description	Default
extra_headers	Headers | None	Send extra headers	None
extra_query	Query | None	Add additional query parameters to the request	None
extra_body	Body | None	Add additional JSON properties to the request	None
timeout	float | Timeout | None | NotGiven	Override the client-level default timeout for this request, in seconds	NOT_GIVEN

with_raw_response

with_raw_response() -> JobsWithRawResponse

with_streaming_response

with_streaming_response() -> JobsWithStreamingResponse

JobsWithRawResponse

JobsWithRawResponse(jobs: Jobs)

Attributes:

Name	Type	Description
cancel
create
list
list_events
retrieve

cancel instance-attribute

cancel = to_raw_response_wrapper(cancel)

create instance-attribute

create = to_raw_response_wrapper(create)

list instance-attribute

list = to_raw_response_wrapper(list)

list_events instance-attribute

list_events = to_raw_response_wrapper(list_events)

retrieve instance-attribute

retrieve = to_raw_response_wrapper(retrieve)

JobsWithStreamingResponse

JobsWithStreamingResponse(jobs: Jobs)

Attributes:

Name	Type	Description
cancel
create
list
list_events
retrieve

cancel instance-attribute

cancel = to_streamed_response_wrapper(cancel)

create instance-attribute

create = to_streamed_response_wrapper(create)

list instance-attribute

list = to_streamed_response_wrapper(list)

list_events instance-attribute

list_events = to_streamed_response_wrapper(list_events)

retrieve instance-attribute

retrieve = to_streamed_response_wrapper(retrieve)