RequestSession Class

Convert to dict, including explicitly set None values.

as_dict(**kwargs: Any) -> dict[str, Any]

Returns

Remove all items from D.

clear() -> None

copy() -> Model

Get the value for key if key is in the dictionary, else default. :param str key: The key to look up. :param any default: The value to return if key is not in the dictionary. Defaults to None :returns: D[k] if k in D, else d. :rtype: any

get(key: str, default: Any = None) -> Any

Parameters

items() -> ItemsView[str, Any]

Returns

keys() -> KeysView[str]

Returns

Removes specified key and return the corresponding value. :param str key: The key to pop. :param any default: The value to return if key is not in the dictionary :returns: The value corresponding to the key. :rtype: any :raises KeyError: If key is not found and default is not given.

pop(key: str, default: ~typing.Any = <object object>) -> Any

Parameters

Removes and returns some (key, value) pair :returns: The (key, value) pair. :rtype: tuple :raises KeyError: if D is empty.

popitem() -> tuple[str, Any]

Same as calling D.get(k, d), and setting D[k]=d if k not found :param str key: The key to look up. :param any default: The value to set if key is not in the dictionary :returns: D[k] if k in D, else d. :rtype: any

setdefault(key: str, default: ~typing.Any = <object object>) -> Any

Parameters

Updates D from mapping/iterable E and F. :param any args: Either a mapping object or an iterable of key-value pairs.

update(*args: Any, **kwargs: Any) -> None

values() -> ValuesView[Any]

Returns

The animation configuration for the session.

animation: '_models.Animation' | None

Configuration for avatar streaming and behavior during the session.

avatar: '_models.AvatarConfig' | None

List of include options for the session (e.g., logprobs, phrases, file search results).

include: list[str | '_models.SessionIncludeOption'] | None

Configuration for echo cancellation during server-side audio processing.

input_audio_echo_cancellation: '_models.AudioEchoCancellation' | None

"pcm16", "g711_ulaw", and "g711_alaw".

input_audio_format: str | '_models.InputAudioFormat' | None

Configuration for input audio noise reduction.

input_audio_noise_reduction: '_models.AudioNoiseReduction' | None

Input audio sampling rate in Hz. Available values:

• For pcm16: 8000, 16000, 24000

• For g711_alaw/g711_ulaw: 8000.

input_audio_sampling_rate: int | None

Configuration for input audio transcription.

input_audio_transcription: '_models.AudioInputTranscriptionOptions' | None

Optional instructions to guide the model's behavior throughout the session.

instructions: str | None

Configuration for interim response generation during latency or tool calls. Is either a StaticInterimResponseConfig type or a LlmInterimResponseConfig type.

interim_response: '_types.InterimResponseConfig' | None

Maximum number of tokens to generate in the response. Default is unlimited. Is either a int type or a Literal["inf"] type.

max_response_output_tokens: int | Literal['inf'] | None

Set of up to 16 key-value pairs that can be attached to the session. This is useful for storing additional information about the session in a structured format, such as tracking IDs, user context, or application-specific labels. These key-value pairs are also included in Foundry resource logs for tracing and diagnostics. Keys can be a maximum of 64 characters long and values can be a maximum of 512 characters long.

metadata: dict[str, str] | None

The modalities to be used in the session.

modalities: list[str | '_models.Modality'] | None

The model for the session.

model: str | None

"pcm16", "pcm16_8000hz", "pcm16_16000hz", "g711_ulaw", and "g711_alaw".

output_audio_format: str | '_models.OutputAudioFormat' | None

Types of timestamps to include in audio response content.

output_audio_timestamp_types: list[str | '_models.AudioTimestampType'] | None

Constrains effort on reasoning for reasoning models. Check model documentation for supported values for each model. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response. Known values are: "none", "minimal", "low", "medium", "high", and "xhigh".

reasoning_effort: str | '_models.ReasoningEffort' | None

0.0 to 1.0. Default is 0.7.

temperature: float | None

Specifies which tools the model is allowed to call during the session. Is either a Union[str, "_models.ToolChoiceLiteral"] type or a ToolChoiceSelection type.

tool_choice: '_types.ToolChoice' | None

Configuration for tools to be used during the session, if applicable.

tools: list['_models.Tool'] | None

Type of turn detection to use.

turn_detection: '_models.TurnDetection' | None

Union[str, "_models.OpenAIVoiceName"], OpenAIVoice, AzureVoice

voice: '_types.Voice' | None