GET

/v1/schemas/openai/chat/realtime/client_secrets

Accept:

kind inputactivity chathash a4543b27…extracted Nov 4, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `fb72772dc5ac65c0457451fcab125a37ffa22b6aea98546e7be3f49396be4442` { "fetchedAt": 1793769007, "extractorVersion": "1", "verifiedAt": null }

Request body

2 properties · 0 required · 27 $defs

property	type	description	constraints
expires_after	object	Configuration for the client secret expiration. Expiration refers to the time after which a client secret will no longer be valid for creating sessions. The session itself may continue after that time once started. A secret can be used to create multiple sessi…
session	Realtime session configuration \| Realtime transcription session configuration	Session configuration to use for the client secret. Choose either a realtime session or a transcription session.

Types

27 $defs

AudioTranscriptionobject

model	string \| enum	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-…
language	string	The language of the input audio. Supplying the input language in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format will improve accuracy and latency.
languages	string[]	Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe` and `gpt-live-transcribe`.	items ≥ 1
keywords	string[]	Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe` and `gpt-live-transcribe`.
prompt	string	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](https://developers.openai.com/api/docs/guides/speech-to-text#prompting). For `gpt-4o-transcribe` models (excluding `gpt-4o-tra…
delay	enum	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.	minimal \| low \| medium \| high \| xhigh

InputFileContentobject

A file input to the model.

type*	enum	The type of the input item. Always `input_file`.	input_file · default: "input_file"
file_id	string	The ID of the file to be sent to the model.	nullable
filename	string	The name of the file to be sent to the model.
file_data	string	The content of the file to be sent to the model.
prompt_cache_breakpoint	PromptCacheBreakpointConfig	Marks the exact end of a reusable prompt prefix. When `ttl` is omitted, the breakpoint inherits the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.
file_url	string	The URL of the file to be sent to the model.	format: uri
detail	enum	The detail level of the file to be sent to the model. Use `auto` to let the system select the detail level; for GPT-5.6 and later models, `auto` uses high-quality rendering, which may increase input token usage. Use `low` for lower-cost rendering, or `high` to…	auto \| low \| high

InputImageContentobject

An image input to the model. Learn about [image inputs](https://developers.openai.com/api/docs/guides/images-vision).

type*	enum	The type of the input item. Always `input_image`.	input_image · default: "input_image"
image_url	string	The URL of the image to be sent to the model. A fully qualified URL or base64 encoded image in a data URL.	format: uri · nullable
file_id	string	The ID of the file to be sent to the model.	nullable
detail*	enum	The detail level of the image to be sent to the model. One of `high`, `low`, `auto`, or `original`. Defaults to `auto`.	low \| high \| auto \| original
prompt_cache_breakpoint	PromptCacheBreakpointConfig	Marks the exact end of a reusable prompt prefix. When `ttl` is omitted, the breakpoint inherits the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.

InputTextContentobject

A text input to the model.

type*	enum	The type of the input item. Always `input_text`.	input_text · default: "input_text"
text*	string	The text input to the model.
prompt_cache_breakpoint	PromptCacheBreakpointConfig	Marks the exact end of a reusable prompt prefix. When `ttl` is omitted, the breakpoint inherits the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.

MCPToolobject

Give the model access to additional tools via remote Model Context Protocol (MCP) servers. [Learn more about MCP](https://developers.openai.com/api/docs/guides/tools-connectors-mcp).

type*	enum	The type of the MCP tool. Always `mcp`.	mcp
server_label*	string	A label for this MCP server, used to identify it in tool calls.
server_url	string	The URL for the MCP server. One of `server_url`, `connector_id`, or `tunnel_id` must be provided.	format: uri
connector_id	enum	Identifier for service connectors, like those available in ChatGPT. One of `server_url`, `connector_id`, or `tunnel_id` must be provided. Learn more about service connectors [here](https://developers.openai.com/api/docs/guides/tools-connectors-mcp#connectors).	connector_dropbox \| connector_gmail \| connector_googlecalendar \| connector_googledrive \| connector_microsoftteams \| connector_outlookcalendar · +2 more · deprecated
tunnel_id	string	The Secure MCP Tunnel ID to use instead of a direct server URL. One of `server_url`, `connector_id`, or `tunnel_id` must be provided.
authorization	string	An OAuth access token that can be used with a remote MCP server, either with a custom MCP server URL or a service connector. Your application must handle the OAuth authorization flow and provide the token here.
server_description	string	Optional description of the MCP server, used to provide more context.
headers	object	Optional HTTP headers to send to the MCP server. Use for authentication or other purposes.	nullable
allowed_tools	string[] \| MCP tool filter	List of allowed tool names or a filter object.	nullable
allowed_callers	CallableToolAllowedCaller[]	The tool invocation context(s).	items ≥ 1 · nullable
require_approval	MCP tool approval filter \| enum	Specify which of the MCP server's tools require approval.	default: "always" · nullable
defer_loading	boolean	Whether this MCP tool is deferred and discovered via tool search.

MCPToolFilterobject

A filter object to specify which tools are allowed.

tool_names	string[]	List of allowed tool names.
read_only	boolean	Indicates whether or not a tool modifies data or is read-only. If an MCP server is [annotated with `readOnlyHint`](https://modelcontextprotocol.io/specification/2025-06-18/schema#toolannotations-readonlyhint), it will match this filter.

NoiseReductionTypeenum

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.

near_field | far_field

Promptobject

Reference to a prompt template and its variables. [Learn more](https://developers.openai.com/api/docs/guides/text?api-mode=responses#version-prompts-in-code).

id*	string	The unique identifier of the prompt template to use.
version	string	Optional version of the prompt template.	nullable
variables	object	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.	nullable

PromptCacheBreakpointConfigobject

Marks the exact end of a reusable prompt prefix. When `ttl` is omitted, the breakpoint inherits the request's `prompt_cache_options.ttl`; the boundary is not rounded to a token block.

mode*	enum	The breakpoint mode. Always `explicit`.	explicit · default: "explicit"
ttl	enum	The minimum lifetime of this cache breakpoint. Defaults to the request TTL.	30m \| 12h

RealtimeAudioFormatsPCM audio format | PCMU audio format | PCMA audio format

RealtimeFunctionToolobject

type	enum	The type of the tool, i.e. `function`.	function
name	string	The name of the function.
description	string	The description of the function, including guidance on when and how to call it, and guidance about what to tell the user when calling (if anything).
parameters	object	Parameters of the function in JSON Schema.

RealtimeNoiseReductionStrengthnumber

Input noise suppression strength. Higher values apply more suppression. Defaults to 1 when noise reduction is first enabled. Omitting strength from a session update preserves the current strength.

0 ≤ n ≤ 1 · default: 1

RealtimeReasoningobject

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

effort

enum

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

minimal | low | medium | high | xhigh · default: "low"

RealtimeReasoningEffortenum

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

minimal | low | medium | high | xhigh · default: "low"

RealtimeSessionCreateRequestGAobject

Realtime session object configuration.

type*	enum	The type of session to create. Always `realtime` for the Realtime API.	realtime
output_modalities	enum[]	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 `a…	default: ["audio"]
model	string \| enum	The Realtime model used for this session.
instructions	string	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", "he…
audio	object	Configuration for input and output audio.
include	enum[]	Additional fields to include in server outputs.
tracing	enum \| Tracing Configuration	Realtime API can write session traces to the [Traces Dashboard](https://platform.openai.com/logs?api=traces). Set to null to disable tracing. Once tracing is enabled for a session, the configuration cannot be modified.	nullable · default: null
tools	RealtimeFunctionTool \| MCPTool[]	Tools available to the model.
tool_choice	enum \| Function tool \| MCP tool	How the model chooses tools. Provide one of the string modes or force a specific function/MCP tool.	default: "auto"
parallel_tool_calls	boolean	Whether the model may call multiple tools in parallel. Only supported by reasoning Realtime models such as `gpt-realtime-2`.
reasoning	RealtimeReasoning	Configuration for reasoning-capable Realtime models such as `gpt-realtime-2`.
max_output_tokens	integer \| enum	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`.
truncation	enum \| Retention ratio truncation	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…
prompt	object	Reference to a prompt template and its variables. [Learn more](https://developers.openai.com/api/docs/guides/text?api-mode=responses#version-prompts-in-code).	nullable

RealtimeTranscriptionSessionCreateRequestGAobject

Realtime transcription session object configuration.

type*	enum	The type of session to create. Always `transcription` for transcription sessions.	transcription
audio	object	Configuration for input and output audio.
include	enum[]	Additional fields to include in server outputs.

RealtimeTruncationenum | Retention ratio truncation

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…

RealtimeTurnDetectionServer VAD | Semantic VAD

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.

nullable

ResponsePromptVariablesobject

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.

nullable

ToolChoiceFunctionobject

Use this option to force the model to call a specific function.

type*	enum	For function calling, the type is always `function`.	function
name*	string	The name of the function to call.

ToolChoiceMCPobject

Use this option to force the model to call a specific tool on a remote MCP server.

type*	enum	For MCP tools, the type is always `mcp`.	mcp
server_label*	string	The label of the MCP server to use.
name	string	The name of the tool to call on the server.	nullable

ToolChoiceOptionsenum

Controls which (if any) tool is called by the model.

none | auto | required

VoiceIdsOrCustomVoicestring | enum | object

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

VoiceIdsSharedstring | enum

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"openai","endpointId":"realtime/client_secrets","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}