Creates a short-lived Realtime client secret that can include session configuration and be used by client applications without exposing the main API key.

Create a client secret with the given session configuration.

Realtime transcription session object configuration.

The type of session to create. Always transcription for transcription sessions.

The model to use for transcription. Current options are whisper-1, gpt-transcribe, gpt-live-transcribe, gpt-4o-mini-transcribe, gpt-4o-mini-transcribe-2025-12-15, gpt-4o-transcribe, gpt-4o-transcribe-diarize, and gpt-realtime-whisper. Use gpt-4o-transcribe-diarize when you need diarization with speaker labels.

The language of the input audio. Supplying the input language in ISO-639-1 (e.g. en) format will improve accuracy and latency.

Possible languages of the input audio, in ISO-639-1 format. Supported by gpt-transcribe and gpt-live-transcribe.

Words or phrases to guide transcription of the input audio. Supported by gpt-transcribe and gpt-live-transcribe.

An optional text to guide the model's style or continue a previous audio segment. For whisper-1, the prompt is a list of keywords. For gpt-4o-transcribe models (excluding gpt-4o-transcribe-diarize), the prompt is a free text string, for example "expect words related to technology". Prompt is not supported with gpt-realtime-whisper in GA Realtime sessions.

Controls how long the model waits before emitting transcription text. Higher values can improve transcription accuracy at the cost of latency. Only supported with gpt-realtime-whisper in GA Realtime sessions.

Type of noise reduction. near_field is for close-talking microphones such as headphones, far_field is for far-field microphones such as laptop or conference room microphones.

Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to null to turn off, in which case the client must manually trigger model response. Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. For gpt-realtime-whisper transcription sessions, turn detection must be set to null; VAD is not supported.

Additional fields to include in server outputs. item.input_audio_transcription.logprobs: Include logprobs for input audio transcription.

Realtime session object configuration.

The type of session to create. Always realtime for the Realtime API.

The set of modalities the model can respond with. It defaults to ["audio"], indicating that the model will respond with audio plus a transcript. ["text"] can be used to make the model respond with text only. It is not possible to request both text and audio at the same time.

The Realtime model used for this session.

The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. Note that the server sets default instructions which will be used if this field is not set and are visible in the session.created event at the start of the session.

The model to use for transcription. Current options are whisper-1, gpt-transcribe, gpt-live-transcribe, gpt-4o-mini-transcribe, gpt-4o-mini-transcribe-2025-12-15, gpt-4o-transcribe, gpt-4o-transcribe-diarize, and gpt-realtime-whisper. Use gpt-4o-transcribe-diarize when you need diarization with speaker labels.

The language of the input audio. Supplying the input language in ISO-639-1 (e.g. en) format will improve accuracy and latency.

Possible languages of the input audio, in ISO-639-1 format. Supported by gpt-transcribe and gpt-live-transcribe.

Words or phrases to guide transcription of the input audio. Supported by gpt-transcribe and gpt-live-transcribe.

An optional text to guide the model's style or continue a previous audio segment. For whisper-1, the prompt is a list of keywords. For gpt-4o-transcribe models (excluding gpt-4o-transcribe-diarize), the prompt is a free text string, for example "expect words related to technology". Prompt is not supported with gpt-realtime-whisper in GA Realtime sessions.

Controls how long the model waits before emitting transcription text. Higher values can improve transcription accuracy at the cost of latency. Only supported with gpt-realtime-whisper in GA Realtime sessions.

Type of noise reduction. near_field is for close-talking microphones such as headphones, far_field is for far-field microphones such as laptop or conference room microphones.

Configuration for turn detection, ether Server VAD or Semantic VAD. This can be set to null to turn off, in which case the client must manually trigger model response. Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech. Semantic VAD is more advanced and uses a turn detection model (in conjunction with VAD) to semantically estimate whether the user has finished speaking, then dynamically sets a timeout based on this probability. For example, if user audio trails off with "uhhm", the model will score a low probability of turn end and wait longer for the user to continue speaking. This can be useful for more natural conversations, but may have a higher latency. For gpt-realtime-whisper transcription sessions, turn detection must be set to null; VAD is not supported.

A built-in voice name or a custom voice reference.

Additional fields to include in server outputs. item.input_audio_transcription.logprobs: Include logprobs for input audio transcription.

Realtime API can write session traces to the Traces Dashboard. Set to null to disable tracing. Once tracing is enabled for a session, the configuration cannot be modified. auto will create a trace for the session with default values for the workflow name, group id, and metadata.

Tools available to the model.

How the model chooses tools. Provide one of the string modes or force a specific function/MCP tool.

Controls which (if any) tool is called by the model. none means the model will not call any tool and instead generates a message. auto means the model can pick between generating a message or calling one or more tools. required means the model must call one or more tools.

For function calling, the type is always function.

The name of the function to call.

For MCP tools, the type is always mcp.

The label of the MCP server to use.

Whether the model may call multiple tools in parallel. Only supported by reasoning Realtime models such as gpt-realtime-2.

Configuration for reasoning-capable Realtime models such as gpt-realtime-2.

Constrains effort on reasoning for reasoning-capable Realtime models such as gpt-realtime-2.

Maximum number of output tokens for a single assistant response, inclusive of tool calls. Provide an integer between 1 and 4096 to limit output tokens, or inf for the maximum available tokens for a given model. Defaults to inf.

When the number of tokens in a conversation exceeds the model's input token limit, the conversation be truncated, meaning messages (starting from the oldest) will not be included in the model's context. A 32k context model with 4,096 max output tokens can only include 28,224 tokens in the context before truncation occurs. Clients can configure truncation behavior to truncate with a lower max token limit, which is an effective way to control token usage and cost. Truncation will reduce the number of cached tokens on the next turn (busting the cache), since messages are dropped from the beginning of the context. However, clients can also configure truncation to retain messages up to a fraction of the maximum context size, which will reduce the need for future truncations and thus improve the cache rate. Truncation can be disabled entirely, which means the server will never truncate but would instead return an error if the conversation exceeds the model's input token limit.

Reference to a prompt template and its variables. Learn more.

The unique identifier of the prompt template to use.

Optional map of values to substitute in for variables in your prompt. The substitution values can either be strings, or other Response input types like images or files.

The type of the input item. Always input_text.

The text input to the model.

Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's prompt_cache_options.ttl; the boundary is not rounded to a token block.

The breakpoint mode. Always explicit.

The type of the input item. Always input_image.

Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's prompt_cache_options.ttl; the boundary is not rounded to a token block.

The breakpoint mode. Always explicit.

The type of the input item. Always input_file.

The name of the file to be sent to the model.

The content of the file to be sent to the model.

Marks the exact end of a reusable prompt prefix. The breakpoint inherits its TTL from the request's prompt_cache_options.ttl; the boundary is not rounded to a token block.

The breakpoint mode. Always explicit.

The URL of the file to be sent to the model.

The generated client secret value.

Expiration timestamp for the client secret, in seconds since epoch.

A new Realtime transcription session configuration. When a session is created on the server via REST API, the session object also contains an ephemeral key. Default TTL for keys is 10 minutes. This property is not present when a session is updated via the WebSocket API.

The set of modalities the model can respond with. To disable audio, set this to ["text"].

The format of input audio. Options are pcm16, g711_ulaw, or g711_alaw.

The model used for transcription. Current options are whisper-1, gpt-transcribe, gpt-live-transcribe, gpt-4o-mini-transcribe, gpt-4o-mini-transcribe-2025-12-15, gpt-4o-transcribe, gpt-4o-transcribe-diarize, and gpt-realtime-whisper.

The language of the input audio.

The possible input audio languages configured for transcription, in ISO-639-1 format.

The prompt configured for input audio transcription, when present.

A Realtime session configuration object.

Unique identifier for the session that looks like sess_1234567890abcdef.

The object type. Always realtime.session.

Expiration timestamp for the session, in seconds since epoch.

Additional fields to include in server outputs. - item.input_audio_transcription.logprobs: Include logprobs for input audio transcription.

The Realtime model used for this session.

The set of modalities the model can respond with. To disable audio, set this to ["text"].

The default system instructions (i.e. system message) prepended to model calls. This field allows the client to guide the model on desired responses. The model can be instructed on response content and format, (e.g. "be extremely succinct", "act friendly", "here are examples of good responses") and on audio behavior (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The instructions are not guaranteed to be followed by the model, but they provide guidance to the model on the desired behavior. Note that the server sets default instructions which will be used if this field is not set and are visible in the session.created event at the start of the session.

The model used for transcription. Current options are whisper-1, gpt-transcribe, gpt-live-transcribe, gpt-4o-mini-transcribe, gpt-4o-mini-transcribe-2025-12-15, gpt-4o-transcribe, gpt-4o-transcribe-diarize, and gpt-realtime-whisper.

The language of the input audio.

The possible input audio languages configured for transcription, in ISO-639-1 format.

The prompt configured for input audio transcription, when present.

Type of noise reduction. near_field is for close-talking microphones such as headphones, far_field is for far-field microphones such as laptop or conference room microphones.

Configuration options for tracing. Set to null to disable tracing. Once tracing is enabled for a session, the configuration cannot be modified. auto will create a trace for the session with default values for the workflow name, group id, and metadata.

Configuration for turn detection. Can be set to null to turn off. Server VAD means that the model will detect the start and end of speech based on audio volume and respond at the end of user speech.

Tools (functions) available to the model.

How the model chooses tools. Options are auto, none, required, or specify a function.

Maximum number of output tokens for a single assistant response, inclusive of tool calls. Provide an integer between 1 and 4096 to limit output tokens, or inf for the maximum available tokens for a given model. Defaults to inf.

The public explanation for this block.

The public continuation instruction.

The public explanation for this block.

The public continuation instruction.