release: 1.84.0 (#2395)

* feat(api): add new realtime and audio models, realtime session options

* chore(api): update type names

* release: 1.84.0

---------

Co-authored-by: stainless-app[bot] <142633134+stainless-app[bot]@users.noreply.github.com>

Author: stainless-app[bot]

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "1.83.0"
+  ".": "1.84.0"
 }

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 111
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-2bcc845d8635bf93ddcf9ee723af4d7928248412a417bee5fc10d863a1e13867.yml
-openapi_spec_hash: 865230cb3abeb01bd85de05891af23c4
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-0205acb1015d29b2312a48526734c0399f93026d4fe2dff5c7768f566e333fd2.yml
+openapi_spec_hash: 1772cc9056c2f6dfb2a4e9cb77ee6343
 config_hash: ed1e6b3c5f93d12b80d31167f55c557c

CHANGELOG.md

@@ -1,5 +1,18 @@
 # Changelog
 
+## 1.84.0 (2025-06-03)
+
+Full Changelog: [v1.83.0...v1.84.0](https://github.com/openai/openai-python/compare/v1.83.0...v1.84.0)
+
+### Features
+
+* **api:** add new realtime and audio models, realtime session options ([0acd0da](https://github.com/openai/openai-python/commit/0acd0da6bc0468c6c857711bc5e77d0bc6d31be6))
+
+
+### Chores
+
+* **api:** update type names ([1924559](https://github.com/openai/openai-python/commit/192455913b38bf0323ddd0e2b1499b114e2111a1))
+
 ## 1.83.0 (2025-06-02)
 
 Full Changelog: [v1.82.1...v1.83.0](https://github.com/openai/openai-python/compare/v1.82.1...v1.83.0)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "1.83.0"
+version = "1.84.0"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"

src/openai/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "openai"
-__version__ = "1.83.0"  # x-release-please-version
+__version__ = "1.84.0"  # x-release-please-version

src/openai/resources/beta/realtime/sessions.py

@@ -54,14 +54,17 @@ def create(
             "gpt-4o-realtime-preview",
             "gpt-4o-realtime-preview-2024-10-01",
             "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-realtime-preview-2025-06-03",
             "gpt-4o-mini-realtime-preview",
             "gpt-4o-mini-realtime-preview-2024-12-17",
         ]
         | NotGiven = NOT_GIVEN,
         output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        speed: float | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         tool_choice: str | NotGiven = NOT_GIVEN,
         tools: Iterable[session_create_params.Tool] | NotGiven = NOT_GIVEN,
+        tracing: session_create_params.Tracing | NotGiven = NOT_GIVEN,
         turn_detection: session_create_params.TurnDetection | NotGiven = NOT_GIVEN,
         voice: Union[
             str, Literal["alloy", "ash", "ballad", "coral", "echo", "fable", "onyx", "nova", "sage", "shimmer", "verse"]
@@ -129,6 +132,10 @@ def create(
           output_audio_format: The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
               For `pcm16`, output audio is sampled at a rate of 24kHz.
 
+          speed: The speed of the model's spoken response. 1.0 is the default speed. 0.25 is the
+              minimum speed. 1.5 is the maximum speed. This value can only be changed in
+              between model turns, not while a response is in progress.
+
           temperature: Sampling temperature for the model, limited to [0.6, 1.2]. For audio models a
               temperature of 0.8 is highly recommended for best performance.
 
@@ -137,6 +144,12 @@ def create(
 
           tools: Tools (functions) available to the model.
 
+          tracing: 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.
+
           turn_detection: 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
@@ -175,9 +188,11 @@ def create(
                     "modalities": modalities,
                     "model": model,
                     "output_audio_format": output_audio_format,
+                    "speed": speed,
                     "temperature": temperature,
                     "tool_choice": tool_choice,
                     "tools": tools,
+                    "tracing": tracing,
                     "turn_detection": turn_detection,
                     "voice": voice,
                 },
@@ -224,14 +239,17 @@ async def create(
             "gpt-4o-realtime-preview",
             "gpt-4o-realtime-preview-2024-10-01",
             "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-realtime-preview-2025-06-03",
             "gpt-4o-mini-realtime-preview",
             "gpt-4o-mini-realtime-preview-2024-12-17",
         ]
         | NotGiven = NOT_GIVEN,
         output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        speed: float | NotGiven = NOT_GIVEN,
         temperature: float | NotGiven = NOT_GIVEN,
         tool_choice: str | NotGiven = NOT_GIVEN,
         tools: Iterable[session_create_params.Tool] | NotGiven = NOT_GIVEN,
+        tracing: session_create_params.Tracing | NotGiven = NOT_GIVEN,
         turn_detection: session_create_params.TurnDetection | NotGiven = NOT_GIVEN,
         voice: Union[
             str, Literal["alloy", "ash", "ballad", "coral", "echo", "fable", "onyx", "nova", "sage", "shimmer", "verse"]
@@ -299,6 +317,10 @@ async def create(
           output_audio_format: The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
               For `pcm16`, output audio is sampled at a rate of 24kHz.
 
+          speed: The speed of the model's spoken response. 1.0 is the default speed. 0.25 is the
+              minimum speed. 1.5 is the maximum speed. This value can only be changed in
+              between model turns, not while a response is in progress.
+
           temperature: Sampling temperature for the model, limited to [0.6, 1.2]. For audio models a
               temperature of 0.8 is highly recommended for best performance.
 
@@ -307,6 +329,12 @@ async def create(
 
           tools: Tools (functions) available to the model.
 
+          tracing: 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.
+
           turn_detection: 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
@@ -345,9 +373,11 @@ async def create(
                     "modalities": modalities,
                     "model": model,
                     "output_audio_format": output_audio_format,
+                    "speed": speed,
                     "temperature": temperature,
                     "tool_choice": tool_choice,
                     "tools": tools,
+                    "tracing": tracing,
                     "turn_detection": turn_detection,
                     "voice": voice,
                 },

src/openai/types/beta/realtime/session.py

@@ -1,11 +1,19 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 from typing import List, Union, Optional
-from typing_extensions import Literal
+from typing_extensions import Literal, TypeAlias
 
 from ...._models import BaseModel
 
-__all__ = ["Session", "InputAudioNoiseReduction", "InputAudioTranscription", "Tool", "TurnDetection"]
+__all__ = [
+    "Session",
+    "InputAudioNoiseReduction",
+    "InputAudioTranscription",
+    "Tool",
+    "Tracing",
+    "TracingTracingConfiguration",
+    "TurnDetection",
+]
 
 
 class InputAudioNoiseReduction(BaseModel):
@@ -59,6 +67,29 @@ class Tool(BaseModel):
     """The type of the tool, i.e. `function`."""
 
 
+class TracingTracingConfiguration(BaseModel):
+    group_id: Optional[str] = None
+    """
+    The group id to attach to this trace to enable filtering and grouping in the
+    traces dashboard.
+    """
+
+    metadata: Optional[object] = None
+    """
+    The arbitrary metadata to attach to this trace to enable filtering in the traces
+    dashboard.
+    """
+
+    workflow_name: Optional[str] = None
+    """The name of the workflow to attach to this trace.
+
+    This is used to name the trace in the traces dashboard.
+    """
+
+
+Tracing: TypeAlias = Union[Literal["auto"], TracingTracingConfiguration]
+
+
 class TurnDetection(BaseModel):
     create_response: Optional[bool] = None
     """
@@ -175,6 +206,7 @@ class Session(BaseModel):
             "gpt-4o-realtime-preview",
             "gpt-4o-realtime-preview-2024-10-01",
             "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-realtime-preview-2025-06-03",
             "gpt-4o-mini-realtime-preview",
             "gpt-4o-mini-realtime-preview-2024-12-17",
         ]
@@ -188,6 +220,14 @@ class Session(BaseModel):
     sampled at a rate of 24kHz.
     """
 
+    speed: Optional[float] = None
+    """The speed of the model's spoken response.
+
+    1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed.
+    This value can only be changed in between model turns, not while a response is
+    in progress.
+    """
+
     temperature: Optional[float] = None
     """Sampling temperature for the model, limited to [0.6, 1.2].
 
@@ -204,6 +244,16 @@ class Session(BaseModel):
     tools: Optional[List[Tool]] = None
     """Tools (functions) available to the model."""
 
+    tracing: Optional[Tracing] = None
+    """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.
+    """
+
     turn_detection: Optional[TurnDetection] = None
     """Configuration for turn detection, ether Server VAD or Semantic VAD.
 
@@ -227,5 +277,5 @@ class Session(BaseModel):
 
     Voice cannot be changed during the session once the model has responded with
     audio at least once. Current voice options are `alloy`, `ash`, `ballad`,
-    `coral`, `echo` `sage`, `shimmer` and `verse`.
+    `coral`, `echo`, `fable`, `onyx`, `nova`, `sage`, `shimmer`, and `verse`.
     """

src/openai/types/beta/realtime/session_create_params.py

@@ -3,7 +3,7 @@
 from __future__ import annotations
 
 from typing import List, Union, Iterable
-from typing_extensions import Literal, TypedDict
+from typing_extensions import Literal, TypeAlias, TypedDict
 
 __all__ = [
     "SessionCreateParams",
@@ -12,6 +12,8 @@
     "InputAudioNoiseReduction",
     "InputAudioTranscription",
     "Tool",
+    "Tracing",
+    "TracingTracingConfiguration",
     "TurnDetection",
 ]
 
@@ -82,6 +84,7 @@ class SessionCreateParams(TypedDict, total=False):
         "gpt-4o-realtime-preview",
         "gpt-4o-realtime-preview-2024-10-01",
         "gpt-4o-realtime-preview-2024-12-17",
+        "gpt-4o-realtime-preview-2025-06-03",
         "gpt-4o-mini-realtime-preview",
         "gpt-4o-mini-realtime-preview-2024-12-17",
     ]
@@ -94,6 +97,14 @@ class SessionCreateParams(TypedDict, total=False):
     sampled at a rate of 24kHz.
     """
 
+    speed: float
+    """The speed of the model's spoken response.
+
+    1.0 is the default speed. 0.25 is the minimum speed. 1.5 is the maximum speed.
+    This value can only be changed in between model turns, not while a response is
+    in progress.
+    """
+
     temperature: float
     """Sampling temperature for the model, limited to [0.6, 1.2].
 
@@ -110,6 +121,16 @@ class SessionCreateParams(TypedDict, total=False):
     tools: Iterable[Tool]
     """Tools (functions) available to the model."""
 
+    tracing: Tracing
+    """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.
+    """
+
     turn_detection: TurnDetection
     """Configuration for turn detection, ether Server VAD or Semantic VAD.
 
@@ -205,6 +226,29 @@ class Tool(TypedDict, total=False):
     """The type of the tool, i.e. `function`."""
 
 
+class TracingTracingConfiguration(TypedDict, total=False):
+    group_id: str
+    """
+    The group id to attach to this trace to enable filtering and grouping in the
+    traces dashboard.
+    """
+
+    metadata: object
+    """
+    The arbitrary metadata to attach to this trace to enable filtering in the traces
+    dashboard.
+    """
+
+    workflow_name: str
+    """The name of the workflow to attach to this trace.
+
+    This is used to name the trace in the traces dashboard.
+    """
+
+
+Tracing: TypeAlias = Union[Literal["auto"], TracingTracingConfiguration]
+
+
 class TurnDetection(TypedDict, total=False):
     create_response: bool
     """