Связаться с намиПодключиться

SDK

Статья создана
Обновлена 7 ноября 2025 г.

class yandex_cloud_ml_sdk.YCloudML

The main class that needs to be instantiated to work with SDK.

tools: Tools

Domain for creating various tools for assistants and function calling

models: Models

Domain for working with models (inference and tuning)

files: Files

Domain for working with files (a part of the Asssistants API)

threads: Threads

Domain for working with threads (a part of the Assistants API)

assistants: Assistants

Domain for working with assistants (a part of the Assistants API)

runs: Runs

Domain for working with assistants’ runs (a part of the Assistants API)

search_api: SearchAPIDomain

Domain for working with Search API

search_indexes: SearchIndexes

Domain for working with search indexes (a part of the Assistants API)

datasets: Datasets

Domain for working with datasets

tuning: Tuning

Domain for working with tuning

batch: Batch

Domain for working with batch tasks

chat: Chat

Domain for working with Yandex Cloud OpenAI Compatible API_BaseSDK_URL.

__init__(*, folder_id=Undefined, endpoint=Undefined, auth=Undefined, retry_policy=Undefined, yc_profile=Undefined, service_map=Undefined, interceptors=Undefined, enable_server_data_logging=Undefined, verify=Undefined)

Construct a new asynchronous sdk instance.

Parameters

  • folder_id (str) – Yandex Cloud folder identifier which will be billed for models usage. In case of default Undefined value, the parameter will be taken from the environment variable YC_FOLDER_ID.
  • endpoint (str) – domain:port pair for Yandex Cloud API or any other grpc compatible target. In case of None passed it turns off service endpoint discovery mechanism and requires service_map to be passed.
  • auth (str | BaseAuth | Undefined) – string with API Key, IAM token or one of yandex_cloud_ml_sdk.auth objects; in case of default Undefined value, the token will be taken from one of the environment variables: YC_OAUTH_TOKEN, YC_TOKEN, YC_IAM_TOKEN, or YC_API_KEY.
  • service_map (Dict[str, str]) – a way to redefine endpoints for one or more cloud subservices with a format of dict {"service_name": "service_address"}.
  • enable_server_data_logging (bool | Undefined) – when passed bool, we will add x-data-logging-enabled: to all of requests, which will enable or disable logging of user data on server side. It will do something only on those parts of backends which supports this option.
  • verify (bool | pathlib.Path | str | os.PathLike) – SSL certificates (a.k.a CA bundle) used to verify the identity of requested hosts. Either True (default CA bundle), a path to an SSL certificate file, or False (which will disable verification).
  • retry_policy (RetryPolicy | Undefined)
  • yc_profile (str | Undefined)
  • interceptors (Sequence[ClientInterceptor] | Undefined)

setup_default_logging(log_level='INFO', log_format='[%(levelname)1.1s %(asctime)s %(name)s:%(lineno)d] %(message)s', date_format='%Y-%m-%d %H:%M:%S')

Sets up the default logging configuration.

Read more about log_levels, log_format, and date_format in Python documentation (logging).

Parameters

  • log_level (Literal['CRITICAL', 'FATAL', 'ERROR', 'WARN', 'WARNING', 'INFO', 'DEBUG', 'TRACE'] | ~typing.Literal['critical', 'fatal', 'error', 'warn', 'warning', 'info', 'debug', 'TRACE'] | int) – The logging level to set.
  • log_format (str) – The format of the log messages.
  • date_format (str) – The format for timestamps in log messages.

Returns

The instance of the SDK with logging configured.

Return type

Self
Предыдущая
Overview
Следующая
Models domain