time multimodal-Server events

Server events for the Qwen-Omni-Realtime API, including function calling events.

See Real-time (Qwen-Omni-Realtime).

error

Indicates a server error message. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_RoUu4T8yExPMI37GKwaOC",
 "type": "error",
 "error": {
 "type": "invalid_request_error",
 "code": "invalid_value",
 "message": "Invalid modalities: \['audio'\]. Supported combinations are: \['text'\] and \['audio', 'text'\].",
 "param": "session.modalities"
 }
}

     **type **`*string*`* * The event type. This value is always `error`.     **error **`*object*`* * Detailed error information.

Properties

**type **string* * The error type.

**code **string* * The error code.

**message ***string* The error message.

**param ***string* The parameter associated with the error, such as session.modalities.

session.created

The server returns this event after a client connects. It contains the default configuration for the session. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_RdvlSpbBb2ssyBjYrDHjt",
 "type": "session.created",
 "session": {
 "object": "realtime.session",
 "model": "qwen3-omni-flash-realtime",
 "modalities": \[
 "text",
 "audio"
 \],
 "voice": "Cherry",
 "input_audio_format": "pcm",
 "output_audio_format": "pcm",
 "input_audio_transcription": {
 "model": "qwen3-asr-flash-realtime"
 },
 "turn_detection": {
 // The value can be server_vad or semantic_vad (only supported by qwen3.5-omni-realtime).
 "type": "server_vad",
 "threshold": 0.5,
 "prefix_padding_ms": 300,
 "silence_duration_ms": 800,
 "create_response": true,
 "interrupt_response": true
 },
 "enable_search": false,
 "search_options": {},
 "tools": \[\],
 "temperature": 0.8,
 "id": "sess_Ov7GOXoNXhNjlxXtOGKQS"
 }
}

     **type **`*string*`* * The event type. This value is always `session.created`.     **session **`*object*`* * The session configuration.

Properties

**object **string* * This value is always realtime.session.

**model **string* * The model used.

**modalities **array* * The output modalities for the model.

**voice **string* * The voice for model-generated audio.

**input_audio_format ***string* The input audio format. This value is always pcm.

**output_audio_format ***string* The output audio format. This value is always pcm.

**input_audio_transcription **object* * The transcription configuration. Properties

**model ***string* The transcription model. This value is always qwen3-asr-flash-realtime. This parameter is not configurable.

**turn_detection **object* * The voice activity detection (VAD) configuration. Properties

**type **string* * The VAD type. Valid values are server_vad (default) or semantic_vad. See Client events.

**threshold **float* * The VAD detection threshold.

**silence_duration_ms **integer* * The duration of silence in milliseconds that triggers the detection of the end of speech.

**enable_search ***boolean* Whether to enable web search. Supported only by Qwen3.5-Omni-Realtime.

**search_options ***object* The options for the web search.

**temperature **float* * The temperature parameter for the model.

session.updated

The server returns this event after a session.update request succeeds. If the request fails, the server returns an error event. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_X1HsXS4b4uptp6yo1LgKd",
 "type": "session.updated",
 "session": {
 "id": "sess_Aih6vAcY5Ddt6jwFx1tCa",
 "object": "realtime.session",
 "model": "qwen3-omni-flash-realtime",
 "modalities": \[
 "text",
 "audio"
 \],
 "instructions": "You are Xiao Yun, a personal assistant. Answer user questions accurately and in a friendly manner. Always respond with a helpful attitude.",
 "voice": "Cherry",
 "input_audio_format": "pcm",
 "output_audio_format": "pcm",
 "input_audio_transcription": {
 "model": "qwen3-asr-flash-realtime"
 },
 "turn_detection": {
 // The value can be server_vad or semantic_vad (only supported by qwen3.5-omni-realtime).
 "type": "server_vad",
 "threshold": 0.1,
 "prefix_padding_ms": 500,
 "silence_duration_ms": 900,
 "create_response": true,
 "interrupt_response": true
 },
 "enable_search": true,
 "search_options": {
 "enable_source": true
 },
 "tools": \[
 {
 "type": "function",
 "function": {
 "name": "get_current_weather",
 "description": "Useful for querying the weather in a specific city.",
 "parameters": {
 "type": "object",
 "properties": {
 "location": {"type": "string", "description": "The city name"}
 },
 "required": \["location"\]
 }
 }
 }
 \],
 "temperature": 0.8,
 "max_response_output_token": "inf",
 "max_tokens": 16384,
 "repetition_penalty": 1.05,
 "presence_penalty": 0.0,
 "top_k": 50,
 "top_p": 1.0,
 "seed":-1
 }
}

     **type **`*string*`* * The event type. This value is always `session.updated`.     **session **`*object*`* * The session configuration.

Properties

**temperature **float* * The temperature parameter for the model.

**modalities **array* * The output modalities for the model.

**voice **string* * The voice for model-generated audio.

**instructions **string* * The model's goal and role.

**input_audio_format ***string* The input audio format. Only pcm is supported.

**output_audio_format ***string* The output audio format. Only pcm is supported.

**input_audio_transcription **object* * The transcription configuration. Properties

**model ***string* The transcription model. This value is always qwen3-asr-flash-realtime. This parameter is not configurable.

**turn_detection **object* * The voice activity detection (VAD) configuration. Properties

**type **string* * The VAD type. Valid values are server_vad (default) or semantic_vad. See Client events.

**threshold **float* * The VAD detection threshold.

**silence_duration_ms **integer* * The duration of silence in milliseconds that triggers the detection of the end of speech.

**enable_search ***boolean* (optional) Whether to enable web search. Supported only by Qwen3.5-Omni-Realtime.

**search_options ***object* (optional) The web search options.

**tools ***array* (optional) A list of tool definitions. When you configure tools, the model can decide whether to call a tool based on the user's input. Properties

**type ***string* (required) This value is always function.

**function.name ***string* (required) The name of the custom tool function. We recommend using the same name as the function, such as get_current_weather or get_current_time.

**function.description ***string* (optional) A description of the tool function's purpose. The model uses this field to decide whether to use the tool function.

**function.parameters ***object* (optional) A description of the tool function's input parameters. The model uses this field to extract the input parameters. If the tool function does not require input parameters, you do not need to specify this parameter. Properties

**type ***string* (required) This value is always object.

**properties ***object* (optional) Describes the name, data type, and description of each input parameter. The key is the parameter name, and the value is an object that contains the data type (type) and description (description).

**required ***array* (optional) Specifies which input parameters are required.

**top_p **float* * The probability threshold for nucleus sampling.

**top_k ***integer* The size of the candidate set for sampling during model generation.

**max_tokens ***integer* The maximum number of tokens that the model can return for the request.

**repetition_penalty **float* * Controls repetition in consecutive sequences during generation.

**presence_penalty **float* * Controls repetition in generated content.

**seed **integer* * The degree of consistency in the model's output per request.

input_audio_buffer.speech_started

In VAD mode, the server returns this event when it detects the start of speech in the audio buffer.

This event may be triggered each time you add audio to the buffer before speech is detected. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_Pvp8nEhsQuGCQbFJ9x58n",
 "type": "input_audio_buffer.speech_started",
 "audio_start_ms": 3647,
 "item_id": "item_YbAiGvK2H7YaS34o4R6Ba"
}

     **type **`*string*`* * The event type. This value is always `input_audio_buffer.speech_started`.     **audio_start_ms **`*integer*`* * The time in milliseconds from when audio writing to the buffer starts until speech is first detected.     **item_id **`*string*`* * The ID of the user message item that is created when the end of speech is detected. ** User message items are used to append user input to the conversation history for subsequent model inference and generation.

input_audio_buffer.speech_stopped

In VAD mode, the server returns this event when it detects the end of speech in the audio buffer.

The server also returns a conversation.item.created event that creates the corresponding user message item. <b>event_id **string* * A unique identifier for this event.

{
 "event_id": "event_UhQiqNVRsgUiq4KUS5Xb5",
 "type": "input_audio_buffer.speech_stopped",
 "audio_end_ms": 4453,
 "item_id": "item_YbAiGvK2H7YaS34o4R6Ba"
}

     **type **`*string*`* * The event type. This value is always `input_audio_buffer.speech_stopped`.     **audio_end_ms **`*integer*`* * The time in milliseconds from the start of the session until the end of speech is detected.     **item_id **`*string*`* * The ID of the user message item that will be created.

input_audio_buffer.committed

The server returns this event when the input audio buffer is committed.

• In VAD mode, the server automatically commits the audio buffer and returns this event when it detects the end of speech.

• In Manual mode, the server returns this event after the client sends an input_audio_buffer.commit event.

  **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_Iy6sUzL1nmdFgshFYxJEz",
 "type": "input_audio_buffer.committed",
 "item_id": "item_YbAiGvK2H7YaS34o4R6Ba"
}

     **type **`*string*`* * The event type. This value is always `input_audio_buffer.committed`.     **item_id **`*string*`* * The ID of the user message item that will be created.

input_audio_buffer.cleared

The server returns this event after the client sends an input_audio_buffer.clear event. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_RoUu4T8yExPMI37GKwaOC",
 "type": "input_audio_buffer.cleared"
}

     **type **`*string*`* * The event type. This value is always `input_audio_buffer.cleared`.

conversation.item.created

The server returns this event when a conversation item is created. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_JEfkrr9gO3Ny7Xcv9bGVd",
 "type": "conversation.item.created",
 "item": {
 "id": "item_YbAiGvK2H7YaS34o4R6Ba",
 "object": "realtime.item",
 "type": "message",
 "status": "in_progress",
 "role": "assistant",
 "content": \[
 {
 "type": "input_audio"
 }
 \]
 }
}
// Tool calling scenario
{
 "event_id": "event_S1hkaIQgcuQD8OEdOpGHQ",
 "type": "conversation.item.created",
 "item": {
 "id": "item_FEG9qJGNkPcdf4et3p7BV",
 "object": "realtime.item",
 "type": "function_call",
 "status": "in_progress",
 "call_id": "call_bc0a7fb7235840f69ecfe4",
 "name": "get_current_weather",
 "arguments": ""
 }
}

     **type **`*string*`* * The event type. This value is always `conversation.item.created`.     **item **`*object*`* * The conversation item to add.

Properties

**id **string* * The unique ID of the conversation item.

**object **string* * This value is always realtime.item.

**status **string* * The status of the conversation item.

**role **string* * The role of the message.

**content **string* * The content of the message. This parameter is returned when the type is message.

**type **string* * The type of the conversation item. Valid values are message or function_call.

**name **string* * The name of the function that is called when the type is function_call.

**call_id **string* * When the type is function_call, this is the unique ID of the function invocation.

**arguments **string* * When the type is function_call, this parameter contains the arguments for the function invocation as a JSON string.

conversation.item.input_audio_transcription.delta

When input audio transcription is enabled, this event is sent frequently while the user is speaking. It provides real-time intermediate transcription results. You can concatenate text + stash to get the most complete sentence preview at any point in time. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_C7jzoeSFuiwOZS6tR14yx",
 "type": "conversation.item.input_audio_transcription.delta",
 "item_id": "item_ThVYhLHOdeXb4bBSvzSFF",
 "content_index": 0,
 "text": "",
 "stash": "How is the weather today?",
 "language": "en",
 "emotion": "neutral",
 "obfuscation": "ABEXGYmxdmc97u"
}

To get the most complete sentence preview at any moment, concatenate these two fields: real-time preview = text + stash. Click to view example Suppose the user is saying: "The weather is nice today, sunny and warm." The following shows the event stream you might receive and how to interpret them:

Time	User speech progress	API response (text and stash)	Client UI display (text + stash)
T1	"The weather..."	text: "" stash: "The weather"	The weather
T2	"...is nice..."	text: "" stash: "The weather is nice"	The weather is nice
T3	"...today,"	text: "The weather" stash: " is nice today,"	The weather is nice today, ("The weather" has been confirmed and moved to text)
T4	(brief pause)	text: "The weather is nice today, " stash: ""	The weather is nice today, (first clause fully confirmed)
T5	"sunny..."	text: "The weather is nice today, " stash: "sunny"	The weather is nice today, sunny
T6	"...and warm."	text: "The weather is nice today, " stash: "sunny and warm."	The weather is nice today, sunny and warm.
T7	(stops speaking)	-	Use the transcript from conversation.item.input_audio_transcription.completed as the final result.

     **type **`*string*`* * The event type. This value is always `conversation.item.input_audio_transcription.delta`.     **item_id **`*string*`* * The ID of the associated conversation item.     **content_index **`*integer*`* * The index of the content part that contains the audio.     **text **`*string*`* * The confirmed text prefix. This is the portion of the current sentence that the model has confirmed and will not change.     **stash **`*string*`* * The preliminary text suffix. This is the temporary draft that follows the confirmed portion, which the model is still processing and may revise.     **language **`*string*`* * The detected language of the recognized audio.     **emotion **`*string*`* * The detected emotion of the recognized audio. Valid values: `neutral`, `happy`, `sad`, `angry`, `surprised`, `disgusted`, `fearful`.

conversation.item.input_audio_transcription.completed

Indicates that the user's audio has been transcribed. The transcription is performed by a built-in speech recognition model (qwen3-asr-flash-realtime). This parameter is not configurable.

The transcribed text from the speech recognition model may differ from the interpretation generated by the Qwen-Omni-Realtime model. The transcription is for reference only. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_FrrZcxiDfTB9LD9p4pVng",
 "type": "conversation.item.input_audio_transcription.completed",
 "item_id": "item_YbAiGvK2H7YaS34o4R6Ba",
 "content_index": 0,
 "transcript": "Hello."
}

     **type **`*string*`* * The event type. This value is always `conversation.item.input_audio_transcription.completed`.     **item_id **`*string*`* * The ID of the user message item.     **content_index ***integer * This value is always 0.     **transcript **`*string*`* * The transcribed text.

conversation.item.input_audio_transcription.failed

The server returns this event when input audio transcription is enabled and the transcription fails. This event is independent of the error event and helps clients identify transcription failures. **event_id **string* * A unique identifier for this event.

{
 "type": "conversation.item.input_audio_transcription.failed",
 "item_id": "*",
 "content_index": 0,
 "error": {
 "code": "<code>",
 "message": "<message>",
 "param": "\"
 }
}

     **type **`<i>string*`* * The event type. This value is always `conversation.item.input_audio_transcription.failed`.     **item_id **`*string*`* * The ID of the user message item.     **content_index **`*integer*`* * This value is always 0.     **error **`*object*`* * The error information.

Properties

code *string** * The error code.

message *string** * The error message.

param *string** * The parameter related to the error.

response.created

The server returns this event when it generates a new model response. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_XuDavMzQN3KKepqGu3KRh",
 "type": "response.created",
 "response": {
 "id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "object": "realtime.response",
 "conversation_id": "conv_FjJaccpnvwHNo9cPVuzGc",
 "status": "in_progress",
 "modalities": \[
 "text",
 "audio"
 \],
 "voice": "Cherry",
 "output_audio_format": "pcm",
 "output": \[\]
 }
}

     **type **`*string*`* * The event type. This value is always `response.created`.     **response **`*object*`* * The response object.

Properties

id *string** * The unique ID of the response.

conversation_id *string** * The unique ID of the current session.

object *string** * The object type. For this event, this value is always realtime.response.

status *string** * The response status. Valid values are completed, failed, in_progress, or incomplete.

modalities *array** * The response modalities.

voice *string** * The voice for model-generated audio.

output *string** * This field is empty for this event.

response.done

The server returns this event after the response is completely generated. The response object includes all output items except for the raw audio data. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_CSaxRRYLvbrfexDXAEuDG",
 "type": "response.done",
 "response": {
 "id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "object": "realtime.response",
 "conversation_id": "conv_FjJaccpnvwHNo9cPVuzGc",
 "status": "completed",
 "modalities": \[
 "text",
 "audio"
 \],
 "voice": "Cherry",
 "output_audio_format": "pcm",
 "output": \[
 {
 "id": "item_Ls6MtCUWO7LM4E59QziNv",
 "object": "realtime.item",
 "type": "message",
 "status": "completed",
 "role": "assistant",
 "content": \[
 {
 "type": "audio",
 "transcript": "Hello! How can I help you?"
 }
 \]
 }
 \],
 "usage": {
 "total_tokens": 377,
 "input_tokens": 336,
 "output_tokens": 41,
 "input_tokens_details": {
 "text_tokens": 228,
 "audio_tokens": 108
 },
 "output_tokens_details": {
 "text_tokens": 9,
 "audio_tokens": 32
 },
 "plugins": {
 "search": {
 "count": 1,
 "strategy": "agent"
 }
 }
 }
 }
}
// Tool calling scenario
{
 "event_id": "event_T1EFAJp43X2DWtDRmxTtx",
 "type": "response.done",
 "response": {
 "id": "resp_TucN5QgymL5MA8vkJvFlS",
 "object": "realtime.response",
 "conversation_id": "conv_SEDZESRlefT8WvLSmEn6E",
 "status": "completed",
 "modalities": \["text", "audio"\],
 "voice": "Ethan",
 "output_audio_format": "pcm16",
 "output": \[
 {
 "id": "item_FEG9qJGNkPcdf4et3p7BV",
 "object": "realtime.item",
 "type": "function_call",
 "status": "completed",
 "call_id": "call_bc0a7fb7235840f69ecfe4",
 "name": "get_current_weather",
 "arguments": " {\\"location\\": \\"Hangzhou\\"}"
 }
 \],
 "usage": {
 "total_tokens": 567,
 "input_tokens": 524,
 "output_tokens": 43,
 "input_tokens_details": {
 "text_tokens": 487,
 "audio_tokens": 37
 },
 "output_tokens_details": {
 "text_tokens": 43
 }
 }
 }
}

     **type **`*string*`* * The event type. This value is always `response.done`.     **response **`*object*`* * The response object.

Properties

id *string** * The unique ID of the response.

conversation_id *string** * The unique ID of the current session.

object *string** * The object type. For this event, this value is always realtime.response.

status *string** * The response status.

modalities *array** * The response modalities.

voice *string** * The voice for model-generated audio.

output *object** * The response output. Properties

id *string** * The ID of the response output.

type *string** * The type of the output item. Valid values are message or function_call.

object *string** * The output item object type. This value is always realtime.item.

status *string** * The output item status.

role *string** * The output item role.

content *array** * The content of the output item. This field is returned only when the type is message. Properties

type *string** * The content type. The value can be text for plain text output or audio for audio output.

text *string** * The text output.

transcript *string** * The text transcript of the audio.

name *string** * The name of the function that is invoked when the type is function_call.

call_id *string** * When the type is function_call, this is the unique ID of the function invocation.

arguments *string** * When the type is function_call, this field contains the full arguments for the function call as a JSON string.

usage *object** * Token usage details for this response. Properties

total_tokens *integer* The total number of tokens used in this response.

input_tokens *integer* The number of input tokens.

output_tokens *integer* The number of output tokens.

input_tokens_details *object* Details about input token usage, including text_tokens and audio_tokens.

output_tokens_details *object* Details about output token usage, including text_tokens and audio_tokens.

plugins *object* (optional) Plugin usage metrics. This field is returned when web search (enable_search) is enabled. Properties

search *object* Search metering data. Properties

count *integer* The number of searches.

strategy *string* The search strategy.

response.text.delta

The server returns this event when the output modality is text only and the model generates new text incrementally. **event_id **string* * A unique identifier for this event.

{
 "delta": "Hello",
 "event_id": "event_TH49MauuPmRo1RGaMSlP7",
 "type": "response.text.delta",
 "response_id": "resp_PrRSvPVpnCExdUOGHHLuP",
 "item_id": "item_L8IRm9kRXFpxoOjDqDC96",
 "output_index": 0,
 "content_index": 0
}

     **type **`*string*`* * The event type. This value is always `response.text.delta`.     **delta **`*string*`* * The returned incremental text.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID. You can use this to reference the same message item.     **output_index **`*integer*`* * The index of the output item in the response. This value is always 0.     **content_index **`*integer*`* * The index of the internal part within the output item. This value is always 0.

response.text.done

When the output modality is text only and the model finishes generating text, the server returns this event.

The server also returns this event if the response is interrupted, incomplete, or canceled. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_B1lIeE2Nac33zn5V7h2mm",
 "type": "response.text.done",
 "response_id": "resp_B1lIdtjF4Noqpn5NOjznj",
 "item_id": "item_B1lIdJsAJlJiFs8ztWpJt",
 "output_index": 0,
 "content_index": 0,
 "text": "How can I assist you today?"
}

     **type **`*string*`* * The event type. This value is always `response.text.done`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **content_index **`*integer*`* * The index of the output item in the response.     **text** `*string*`* * The full text generated by the model.

response.audio.delta

When the output modality includes audio and the model generates new audio data incrementally, the server returns this event. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_B1osWMZBtrEQbiIwW0qHQ",
 "type": "response.audio.delta",
 "response_id": "resp_P79OOMs8LnrXVpiIHUCKR",
 "item_id": "item_OFaPGtzfWCPyGzxnuEX9i",
 "output_index": 0,
 "content_index": 0,
 "delta": "{base64 audio}"
}

     **type **`*string*`* * The event type. This value is always `response.audio.delta`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **content_index **`*integer*`* * The index of the output item in the response.     **delta** `*string*` The incremental audio data generated by the model, which is Base64-encoded.

response.audio.done

When the output modality includes audio and the model finishes generating audio data, the server returns this event.

The server also returns this event if the response is interrupted, incomplete, or canceled. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_Le1TDl7VfyHQxl47DtGxI",
 "type": "response.audio.done",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "item_id": "item_Ls6MtCUWO7LM4E59QziNv",
 "output_index": 0,
 "content_index": 0
}

     **type **`*string*`* * The event type. This value is always `response.audio.done`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **content_index **`*integer*`* * The index of the output item in the response.

response.audio_transcript.delta

The server returns a response.audio_transcript.delta event when the output modality includes audio and the model generates new text for the audio incrementally. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_BksW7fOwnyavZdDxIzZYM",
 "type": "response.audio_transcript.delta",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "item_id": "item_Ls6MtCUWO7LM4E59QziNv",
 "output_index": 0,
 "content_index": 0,
 "delta": "What"
}

     **type **`*string*`* * The event type. This value is always `response.audio_transcript.delta`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **content_index **`*integer*`* * The index of the output item in the response.     **delta **`*string*`* * The incremental text.

response.audio_transcript.done

The server returns a response.audio_transcript.done event when the output modality includes audio and the model completes transcribing the audio. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_X49tL2WerT4WjxcmH16lS",
 "type": "response.audio_transcript.done",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "item_id": "item_Ls6MtCUWO7LM4E59QziNv",
 "output_index": 0,
 "content_index": 0,
 "transcript": "Hello! How can I help you?"
}

     **type **`*string*`* * The event type. This value is always `response.audio_transcript.done`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **content_index **`*integer*`* * The index of the output item in the response.     **transcript **`*string*`* * The full transcription text.

response.function_call_arguments.delta

When the model generates the argument string for a function call in a streaming manner, the server pushes this event for each new segment of content. You should concatenate the delta fields from each event in the order they are received to obtain the current argument text. The complete content is provided in the subsequent response.function_call_arguments.done event. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_SlKoJyEbPEqLq14DSM1u5",
 "type": "response.function_call_arguments.delta",
 "response_id": "resp_JnTOsWXlFhKcFohZbtfz6",
 "item_id": "item_Rhcms7CauTNsQprV5S4Hr",
 "output_index": 0,
 "call_id": "call_2be200f4cafe419b9530dd",
 "delta": " {\\"location\\": \\"Beijing\\"}"
}

     **type **`*string*`* * The event type. This value is always `response.function_call_arguments.delta`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **call_id **`*string*`* * The unique ID for this function invocation. This is consistent with the `done` event in the same turn.     **delta **`*string*`* * The new segment of the argument string (delta). You must concatenate these segments in order.

response.function_call_arguments.done

Indicates that the function call arguments have been fully generated. The arguments field in this event contains the complete argument string. After you receive this event, you can parse the arguments and call the local tool function. You must use the complete arguments from this event, not the concatenated delta result. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_X6suLyuL5agdH7r6koesM",
 "type": "response.function_call_arguments.done",
 "response_id": "resp_JnTOsWXlFhKcFohZbtfz6",
 "item_id": "item_Rhcms7CauTNsQprV5S4Hr",
 "output_index": 0,
 "name": "get_current_weather",
 "call_id": "call_2be200f4cafe419b9530dd",
 "arguments": " {\\"location\\": \\"Beijing\\"}"
}

     **type **`*string*`* * The event type. This value is always `response.function_call_arguments.done`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response.     **call_id **`*string*`* * The unique ID for this function invocation.     **name **`*string*`* * The name of the function that was called.     **arguments **`*string*`* * The complete arguments for the function invocation, typically represented as a JSON string.

response.output_item.added

The server returns this event when it creates a new item during response generation. The item type can be message or function_call. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_DsCO341DEVtiATtCB6BUY",
 "type": "response.output_item.added",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "output_index": 0,
 "item": {
 "id": "item_Ls6MtCUWO7LM4E59QziNv",
 "object": "realtime.item",
 "type": "message",
 "status": "in_progress",
 "role": "assistant",
 "content": \[\]
 }
}
// Tool calling scenario
{
 "event_id": "event_HXmKt5pGoiRtXx7Hq7zpN",
 "type": "response.output_item.added",
 "response_id": "resp_TucN5QgymL5MA8vkJvFlS",
 "output_index": 0,
 "item": {
 "id": "item_FEG9qJGNkPcdf4et3p7BV",
 "object": "realtime.item",
 "type": "function_call",
 "status": "in_progress",
 "call_id": "call_bc0a7fb7235840f69ecfe4",
 "name": "get_current_weather",
 "arguments": ""
 }
}

     **type **`*string*`* * The event type. This value is always `response.output_item.added`.     **response_id **`*string*`* * The response ID.     **output_index **`*integer*`* * The index of the output item in the response.     **item **`*object*`* * Information about the output item.

Properties

**id **string* * The unique ID of the output item.

**object **string* * This value is always realtime.item.

**status **string* * The status of the output item.

**role **string* * The role of the sender.

**content **string* * The content of the message. This field is returned when the type is message.

**type **string* * The type of the output item. Valid values are message or function_call.

**name **string* * The name of the function to call when the type is function_call.

**call_id **string* * The unique ID of the current function invocation when the type is function_call.

**arguments **string* * The arguments of the function call as a JSON string when the type is function_call. In an added event, this field is initially an empty string.

response.output_item.done

The server returns this event when the generation of a new output item is complete. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_MEu5nlLw1LsOguHiehIP8",
 "type": "response.output_item.done",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "output_index": 0,
 "item": {
 "id": "item_Ls6MtCUWO7LM4E59QziNv",
 "object": "realtime.item",
 "type": "message",
 "status": "completed",
 "role": "assistant",
 "content": \[
 {
 "type": "audio",
 "text": "Hello! How can I help you?"
 }
 \]
 }
}
// Tool calling scenario
{
 "event_id": "event_FHspdfAnCyjuME3mmAwSY",
 "type": "response.output_item.done",
 "response_id": "resp_TucN5QgymL5MA8vkJvFlS",
 "output_index": 0,
 "item": {
 "id": "item_FEG9qJGNkPcdf4et3p7BV",
 "object": "realtime.item",
 "type": "function_call",
 "status": "completed",
 "call_id": "call_bc0a7fb7235840f69ecfe4",
 "name": "get_current_weather",
 "arguments": " {\\"location\\": \\"Hangzhou\\"}"
 }
}

     **type **`*string*`* * The event type. This value is always `response.output_item.done`.     **response_id **`*string*`* * The response ID.     **output_index **`*integer*`* * The index of the output item in the response.     **item **`*object*`* * The output item information.

Properties

**id **string* * The unique ID of the output item.

**object **string* * This value is always realtime.item.

**status **string* * The status of the output item.

**role **string* * The role of the sender.

**content **string* * The content of the message. This field is returned when the type is message.

**type **string* * The type of the output item. Valid values are message or function_call.

**name **string* * The name of the function that is called when the type is function_call.

**call_id **string* * When the type is function_call, this is the unique ID for the function invocation.

**arguments **string* * When the type is function_call, this field is a JSON string that contains the full arguments for the function call.

response.content_part.added

The server returns this event when it adds a new content part to an assistant message item during response generation. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_AVBOmrgY3C8bjlRajfSUT",
 "type": "response.content_part.added",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "item_id": "item_Ls6MtCUWO7LM4E59QziNv",
 "output_index": 0,
 "content_index": 0,
 "part": {
 "type": "audio",
 "text": ""
 }
}

     **type **`*string*`* * The event type. This value is always `response.content_part.added`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response. This value is always 0.     **content_index **`*integer*`* * The index of the internal part within the output item. This value is always 0.     **part **`*object*`* * The output item information.

Properties

**type **string* * The type of the content part.

**text **string* * The text of the content part.

response.content_part.done

The server returns this event when the streaming of a content part within an assistant message item finishes. **event_id **string* * A unique identifier for this event.

{
 "event_id": "event_Il8HD19v58Qr5IBkw7LtN",
 "type": "response.content_part.done",
 "response_id": "resp_HaVOPdbmX6vifiV5pAfJY",
 "item_id": "item_Ls6MtCUWO7LM4E59QziNv",
 "output_index": 0,
 "content_index": 0,
 "part": {
 "type": "audio",
 "text": "Hello! How can I help you?"
 }
}

     **type **`*string*`* * The event type. This value is always `response.content_part.done`.     **response_id **`*string*`* * The response ID.     **item_id **`*string*`* * The message item ID.     **output_index **`*integer*`* * The index of the output item in the response. This value is always 0.     **content_index **`*integer*`* * The index of the content part in the content array. This value is always 0.     **part **`*object*`* * The output item information.

Properties

**type **string* * The type of the content part.

**text **string* * The text of the content part.