Howie Leung (howieleung) · Nov 9, 2026 · Nov 9, 2026 · Nov 9, 2026 · Nov 9, 2026 · Nov 9, 2026
@@ -1,16 +1,79 @@
 # Release History
 
-## 2.11.0 (2026-11-10)
+## 2.11.0 (2026-11-11)
 
 ### Features Added
 
+* Voice Agents are generally available without preview opt-in for Realtime sessions, stored Realtime conversations, agent generation, and Teams Phone Extensibility/Twilio telephony and call jobs. Use `.voice_agents.realtime`, `.voice_agents.conversations`, `.voice_agents.telephony`, and `.agents.create_from_prompt()`. GPT Live, inline subagent handoff, SIP, post-call delivery, and campaigns remain in preview.
+* Added preview `.beta.voice_agents.live.connect()` and `VoiceAgentLiveDefinition` for GPT Live sessions, with typed session, client/server event, tool, decision, and usage models, session updates, text and audio input, and Responses backend function-tool support.
+* Added Live delegation models for Responses backends, application-managed work, and same-project prompt agents: `VoiceAgentLiveResponsesDelegation`, `VoiceAgentLiveClientDelegation`, and `VoiceAgentLiveManagedAgentDelegation`.
+* Added `.beta.voice_agents.live_conversations` for listing, retrieving, and deleting stored Live conversations, retrieving conversation items, and retrieving or downloading item-level and conversation-level audio.
+* Added optional preview `VoiceAgentDefinition.post_call` with `VoiceAgentPostCallConfig`, `VoiceAgentEventGridPostCallDestination`, and `VoiceAgentRoutinePostCallDestination` for post-call transcript and optional audio delivery to Event Grid connections and routines.
+* Added preview SIP telephony provider support through `TelephonyProvider.SIP`, `SipTelephonyBinding`, and `SipTelephonyBindingListItem`. Added optional `UpdateTelephonyBindingRequest.allowed_calling_numbers` for incoming caller restrictions.
+* Added `.voice_agents.telephony.list_call_jobs()` for listing outbound call jobs with optional status and campaign filters, and `.voice_agents.telephony.delete_call_job()` for deleting a call job.
+* Added preview outbound campaign management through `.beta.voice_agents.telephony`, including long-running creation, cancellation, and deletion, campaign retrieval and listing, validation errors, and operation status. Campaign mutations support ETag match conditions.
+* Added campaign configuration, scheduling, and progress models, including `TelephonyCampaign`, `TelephonyRecipient`, `ImmediateSchedule`, `AtTimeSchedule`, `PerRecipientSchedule`, and `TelephonyCampaignProgress`. `TelephonyCampaignDatasetReference.version` optionally selects a dataset version; when omitted, the latest version is pinned during campaign preparation. Added optional read-only `TelephonyCallJob.campaign_name`.
+* Added preview inline Voice Agent subagents and handoff configuration through `VoiceAgentInlineSubagentConfig`, `InlineVoiceSubagent`, and `VoiceAgentHandoffPolicy`, with handoff lifecycle events and optional `VoiceAgentSessionResponseConfig.handoff`.
+* Added Voice Agent system tools `VoiceAgentTransferSystemTool`, `VoiceAgentSendDtmfSystemTool`, `VoiceAgentSendSmsSystemTool`, `VoiceAgentCollectDtmfSystemTool`, `VoiceAgentCollectSmsSystemTool`, `VoiceAgentDigitRecognitionSystemTool`, and `VoiceAgentLanguageDetectionSystemTool`.
+* Added `VoiceAgentTransferTarget`, `VoiceAgentTransferDestination`, and `VoiceAgentTransferRouting` for configuring system-tool transfer targets and routing.
+* Added optional `VoiceAgentSystemTool.timeout_seconds`, also available on `VoiceAgentEndConversationSystemTool`, for configuring execution timeouts.
+* Added realtime system-tool intent items and argument-streaming and input-collection events through `VoiceAgentSystemToolItem`, `VoiceAgentServerEventSystemToolArgumentsDelta`, `VoiceAgentServerEventSystemToolArgumentsDone`, and `VoiceAgentServerEventInputCollectionStatus`.
+* Added `VoiceAgentBaseDefinition`, `VoiceAgentVoiceConfig`, and `VoiceAgentSubProtocol` for shared voice configuration and Realtime/Live definition types. Added enums for tool types, greetings, interim responses, conversation engines, RTC call errors, and voice-specific event discriminators.
+* Added optional error details `MisalignmentErrorDetailsResource.review_target` and `RealtimeResponseStatusDetailsError.message`.
+* Expanded `ImageGenTool.model` known-value annotations to include the `gpt-image-2.5-sunburst` and `gpt-image-2.5-flare` model families and their dated versions, and `ImageGenTool.quality` to include `xhigh` and `max`.
+* Added explicit recursive typing for nested `CompoundFilter` instances in `CompoundFilter.filters`.
+
+### Breaking Changes
+
+All breaking changes affect previously preview APIs.
+
+Breaking changes in beta methods:
+
+* Moved `.beta.agents.create_from_prompt()` to `.agents.create_from_prompt()`, and `.beta.voice_agents.realtime`, `.beta.voice_agents.conversations`, and the existing `.beta.voice_agents.telephony` binding, call, transfer-target, and call-job methods to `.voice_agents`. GPT Live and campaign management remain under `.beta.voice_agents`.
+* Argument `telephony_binding` renamed to `binding` on the promoted `.voice_agents.telephony.create_binding()` method; the typed body now accepts `TelephonyBinding` instead of `CreateTelephonyBindingRequest`.
+* Argument `body` renamed to `job` on the promoted `.voice_agents.telephony.create_call_job()` method; the typed body now accepts `TelephonyCallJobCreate` instead of `CreateTelephonyCallJobRequest`.
+* Removed the `status` filter from the promoted `.voice_agents.telephony.list_bindings()` method.
+
+Breaking changes in beta classes:
+
+* Removed `CreateTelephonyBindingRequest`, `CreateTeamsPhoneExtensibilityTelephonyBindingRequest`, and `CreateTwilioTelephonyBindingRequest`. Use `TelephonyBinding`, `TeamsPhoneExtensibilityTelephonyBinding`, and `TwilioTelephonyBinding`, respectively.
+* Class `CreateTelephonyCallJobRequest` renamed to `TelephonyCallJobCreate`.
+* Removed `TelephonyOutboundDestination` and `TelephonyOutboundDestinationType`; use `TelephonyAddress` and `TelephonyAddressType`. The `source` and `destination` properties on `TelephonyCallJobCreate` and `TelephonyCallJob` now use `TelephonyAddress`; `source` no longer accepts a plain string.
+* Removed `TelephonyTransferDestination`, `PSTNTelephonyTransferDestination`, `SipTelephonyTransferDestination`, `TeamsTelephonyTransferDestination`, and `TelephonyTransferDestinationKind`. `TelephonyTransferTarget.destination` now uses `TelephonyAddress`, with `PhoneNumber`, `SipAddress`, and `TeamsAddress` variants and `TelephonyAddressType`. The discriminator is `type` instead of `kind`, and the PSTN value is `phone_number` instead of `pstn`.
+* Class `TelephonyOutboundFixedIntervalRetryPolicy` renamed to `FixedIntervalRetryPolicy`.
+* Removed `TelephonyBindingStatus` and the `status` property from `TelephonyBinding`, `TelephonyBindingListItem`, their existing provider-specific subclasses, and `UpdateTelephonyBindingRequest`.
+* Properties `TelephonyBinding.id` and `TelephonyBinding.incoming_call_url` are now read-only, including inherited properties on provider-specific bindings.
+* Removed `TelephonyCallJob.object`. Properties `id`, `agent_name`, `status`, `cancellation`, `attempt_count`, `next_attempt_at`, `terminal_reason`, `revision`, `created_at`, and `updated_at` are now read-only.
+* `VoiceAgentSubagentConfig` is now a polymorphic base requiring `subagent_type` instead of `subagents`. Use `VoiceAgentDelegateSubagentConfig` for the previous referenced-agent configuration, or `VoiceAgentInlineSubagentConfig` for inline subagents.
+* `VoiceAgentResponseCreateParams.input` now uses dedicated response-input models (`RealtimeResponseInputMessage`, `RealtimeResponseInputFunctionCall`, `RealtimeResponseInputFunctionCallOutput`, and `RealtimeResponseItemReference`) instead of `RealtimeConversationItem`.
+* Constructor argument `timestamp_type` on `VoiceAgentServerEventResponseAudioTimestampDelta` is now required; use `VoiceAgentAudioTimestampType.WORD`.
+* Constructor arguments `left` and `right` on `VoiceRecordingChannelLayout` are now required; use `VoiceAudioRole.USER` and `VoiceAudioRole.AGENT`, respectively.
+* Renamed `VoiceAgentInputTranscriptionModel.GPT4_O_TRANSCRIBE`, `GPT4_O_MINI_TRANSCRIBE`, `GPT4_O_TRANSCRIBE_DIARIZE`, and `WHISPER1` to `GPT_4_O_TRANSCRIBE`, `GPT_4_O_MINI_TRANSCRIBE`, `GPT_4_O_TRANSCRIBE_DIARIZE`, and `WHISPER_1`, respectively. Wire values are unchanged.
+* Renamed `ActivityProtocolAccessBoundary.READ1_ON1_*` and `WRITE1_ON1_*` members to `READ_1_ON1_*` and `WRITE_1_ON1_*`, respectively, for the `MANAGER`, `DEVELOPERS`, `ALLOWLISTED`, and `TENANT` boundaries. Wire values are unchanged.
+
+### Bugs Fixed
+
+### Sample updates
+
+* Updated Voice Agent samples to use the GA operation paths without `allow_preview=True`; GPT Live samples retain preview opt-in.
+* Added GPT Live samples for session lifecycle and settings updates, text and audio conversations, local function tools, and same-project managed-agent delegation.
+* Added `sample_voice_agent_live_read_conversation.py` demonstrating stored Live conversation, transcript, and audio retrieval, with an optional temporary greeting session.
+* Added `sample_voice_agent_read_generated_audio.py` demonstrating retrieval of generated, unheard audio from an interrupted Realtime reply.
+* Added a Realtime interruption sample demonstrating response cancellation and conversation-item truncation.
+* Updated existing Voice Agent sample descriptions, configuration guidance, and cleanup instructions.
+* Consolidated the Realtime text-conversation and function-tool examples into their asynchronous samples, and removed the separate asynchronous basic agent sample.
+* Updated multi-agent workflow samples to read workflow action details from serialized streaming event items.
+
+### Features Added
+
+* Voice agents are generally available without preview opt-in for Realtime sessions, stored Realtime conversations, agent generation, and Teams Phone Extensibility/Twilio telephony and call jobs. Use `.voice_agents.realtime`, `.voice_agents.conversations`, `.voice_agents.telephony`, and `.agents.create_from_prompt()`. GPT Live, inline subagent handoff, SIP, post-call delivery, and campaigns remain in preview.
 * Added preview `.beta.voice_agents.live.connect()` and `VoiceAgentLiveDefinition` for synchronous and asynchronous GPT Live sessions, with typed client and server events, session updates, text and audio input, and Responses backend function-tool support.
 * Added Live delegation models for Responses backends, application-managed work, and same-project prompt agents: `VoiceAgentLiveResponsesDelegation`, `VoiceAgentLiveClientDelegation`, and `VoiceAgentLiveManagedAgentDelegation`.
 * Added `.beta.voice_agents.live_conversations` for listing, retrieving, and deleting stored Live conversations, retrieving conversation items, and retrieving or downloading item-level and conversation-level audio.
-* Added optional `VoiceAgentDefinition.post_call` with `VoiceAgentPostCallConfig`, `VoiceAgentEventGridPostCallDestination`, and `VoiceAgentRoutinePostCallDestination` for post-call transcript delivery to Event Grid connections and routines. Audio delivery is not supported in this preview.
+* Added optional preview `VoiceAgentDefinition.post_call` with `VoiceAgentPostCallConfig`, `VoiceAgentEventGridPostCallDestination`, and `VoiceAgentRoutinePostCallDestination` for post-call transcript and audio delivery to Event Grid connections and routines.
 * Added SIP telephony provider support through `TelephonyProvider.SIP`, `SipTelephonyBinding`, and `SipTelephonyBindingListItem`. Added optional `UpdateTelephonyBindingRequest.allowed_calling_numbers` for incoming caller restrictions.
-* Added `.beta.voice_agents.telephony.list_call_jobs()` for listing outbound call jobs with optional status and campaign filters, and `.beta.voice_agents.telephony.delete_call_job()` for deleting a call job.
-* Added `.beta.voice_agents.campaigns` for managing outbound campaigns, including long-running creation, cancellation, and deletion, campaign retrieval and listing, validation errors, and operation status. Campaign mutations support ETag match conditions.
+* Added `.voice_agents.telephony.list_call_jobs()` for listing outbound call jobs with optional status and campaign filters, and `.voice_agents.telephony.delete_call_job()` for deleting a call job.
+* Added `.beta.voice_agents.telephony` for managing outbound campaigns, including long-running creation, cancellation, and deletion, campaign retrieval and listing, validation errors, and operation status. Campaign mutations support ETag match conditions.
 * Added campaign configuration, scheduling, and progress models, including `TelephonyCampaign`, `TelephonyRecipient`, `ImmediateSchedule`, `AtTimeSchedule`, `PerRecipientSchedule`, and `TelephonyCampaignProgress`. `TelephonyCampaignDatasetReference` supports an optional version that pins the latest dataset version during campaign preparation when omitted. Added optional read-only `TelephonyCallJob.campaign_name`.
 * Added inline Voice Agent subagents and handoff configuration through `VoiceAgentInlineSubagentConfig`, `InlineVoiceSubagent`, and `VoiceAgentHandoffPolicy`, with handoff lifecycle events and optional `VoiceAgentSessionResponseConfig.handoff`.
 * Added Voice Agent system tools `VoiceAgentTransferSystemTool`, `VoiceAgentSendDtmfSystemTool`, `VoiceAgentSendSmsSystemTool`, `VoiceAgentCollectDtmfSystemTool`, `VoiceAgentCollectSmsSystemTool`, `VoiceAgentDigitRecognitionSystemTool`, and `VoiceAgentLanguageDetectionSystemTool`.
@@ -27,6 +90,7 @@ All breaking changes affect preview APIs.
 
 Breaking changes in beta methods:
 
+* Moved `.beta.agents.create_from_prompt()` to `.agents.create_from_prompt()`, and `.beta.voice_agents.realtime`, `.beta.voice_agents.conversations`, and `.beta.voice_agents.telephony` to `.voice_agents`. GPT Live and campaigns remain under `.beta.voice_agents`.
 * Argument `telephony_binding` in `.beta.voice_agents.telephony.create_binding()` renamed to `binding`; the typed body now accepts `TelephonyBinding` instead of `CreateTelephonyBindingRequest`.
 * Argument `body` in `.beta.voice_agents.telephony.create_call_job()` renamed to `job`; the typed body now accepts `TelephonyCallJobCreate` instead of `CreateTelephonyCallJobRequest`.
 * Removed the `status` filter from `.beta.voice_agents.telephony.list_bindings()`.
@@ -48,8 +112,13 @@ Breaking changes in beta classes:
 
 ### Bugs Fixed
 
+* Fixed synchronous and asynchronous client imports and campaign polling customizations after the generated campaign operation group moved to `.beta.voice_agents.telephony`.
+* Corrected the package development-status classifier for the stable release.
+
 ### Sample updates
 
+* Corrected workflow streaming sample type narrowing and access to Foundry-specific action fields without changing the streamed output.
+* Updated Voice Agent samples to use the GA operation paths without `allow_preview=True`; GPT Live samples retain preview opt-in.
 * Added Live voice-agent samples under `samples/agents/voice/` for session lifecycle, text and audio conversations, local function tools, and same-project managed-agent delegation.
 * Added `sample_voice_agent_live_read_conversation.py` demonstrating stored Live conversation, transcript, and audio retrieval, with an optional temporary greeting session.
 * Added `sample_voice_agent_read_generated_audio.py` demonstrating retrieval of generated, unheard audio from an interrupted Realtime reply.

@@ -1,7 +1,7 @@
 include *.md
 include LICENSE
 include azure/ai/projects/py.typed
-recursive-include tests *
-recursive-include samples *
+recursive-include tests *.py
+recursive-include samples *.py *.md
 include azure/__init__.py
 include azure/ai/__init__.py
@@ -4,7 +4,7 @@ The AI Projects client library is part of the Microsoft Foundry SDK, and provide
 resources in your [Microsoft Foundry](https://ai.azure.com/) Project. Use it to:
 
 * **Create and run Agents** using methods on the `.agents` client property. This includes **Hosted Agents**, which let you run your own containerized agent runtime while using Microsoft Foundry for managed hosting and scaling.
-* **Build and run Voice Agents (preview)** for real-time, speech-to-speech conversational AI, reachable over a WebSocket (`.beta.voice_agents.realtime`) or telephony (`.beta.voice_agents.telephony`), with persisted conversation transcripts and audio through `.beta.voice_agents.conversations`.
+* **Build and run Voice Agents** for real-time, speech-to-speech conversational AI, reachable over a WebSocket (`.voice_agents.realtime`) or telephony (`.voice_agents.telephony`), with persisted conversation transcripts and audio through `.voice_agents.conversations`. Generate agents from a prompt with `.agents.create_from_prompt()`. These generally available features do not require preview opt-in.
 * **Enhance Agents with specialized tools and toolbox tools** such as:
   * Agent-to-Agent (A2A)
   * Azure AI Search
@@ -45,6 +45,8 @@ resources in your [Microsoft Foundry](https://ai.azure.com/) Project. Use it to:
 
 **Important:** This stable package includes Preview features. Preview features available through stable methods require setting `allow_preview=True` when constructing the client, while Preview features exposed via the .beta sub-clients do not. Preview features may change or be removed in future versions and are not recommended for production use.
 
+GPT Live sessions and stored Live conversations remain under `.beta.voice_agents.live` and `.beta.voice_agents.live_conversations`; outbound campaigns are available under `.beta.voice_agents.telephony`. Campaign operations automatically include the required preview header, while GA telephony bindings and call jobs use `.voice_agents.telephony`. GPT Live definitions, inline subagent handoff, SIP telephony, and post-call delivery remain in preview. Set `allow_preview=True` when creating definitions or using these conditional preview features through `.voice_agents`.
+
 The client library uses version `v1` of the Microsoft Foundry [data plane REST APIs](https://aka.ms/azsdk/azure-ai-projects-v2/api-reference-v1).
 
 [Microsoft Foundry](https://ai.azure.com/)
@@ -195,17 +197,17 @@ The table below lists the operation groups supported by the client library, with
 | Sessions | [Manage hosted sessions](https://learn.microsoft.com/azure/foundry/agents/how-to/manage-hosted-sessions?pivots=python) | `samples/hosted_agents/` |
 | Skills (preview) | | `samples/skills/` |
 | Toolboxes | [Curate intent-based toolbox in Foundry](https://learn.microsoft.com/azure/foundry/agents/how-to/tools/toolbox?pivots=python) | `samples/hosted_agents/`, `samples/toolboxes/` |
-| Voice agents (preview) | | `samples/agents/voice/` |
+| Voice agents | | `samples/agents/voice/` |
 
-### Voice Agent realtime transport (preview)
+### Voice Agent realtime transport
 
-Use `project_client.beta.voice_agents.realtime` for agents created with `VoiceAgentDefinition`.
+Use `project_client.voice_agents.realtime` for agents created with `VoiceAgentDefinition`.
 Connections use WebSocket by default. You can also select the transport explicitly:
 
 ```python
 from azure.ai.projects.models import VoiceAgentTransport
 
-with project_client.beta.voice_agents.realtime.connect(
+with project_client.voice_agents.realtime.connect(
     agent_name="my-voice-agent",
     transport=VoiceAgentTransport.WEBSOCKET,
 ) as connection:
@@ -215,7 +217,7 @@ with project_client.beta.voice_agents.realtime.connect(
 The asynchronous client uses the same keyword:
 
 ```python
-async with async_project_client.beta.voice_agents.realtime.connect(
+async with async_project_client.voice_agents.realtime.connect(
     agent_name="my-voice-agent",
     transport=VoiceAgentTransport.WEBSOCKET,
 ) as connection: