Azure SDK for Python (October 2025)
The Azure SDK team is pleased to announce our October 2025 client library releases.
32 packages released this month.
Stable Packages (14)
-
AI Evaluation
-
Azure AI Search
-
Resource Management - Cognitive Services
-
Resource Management - Compute
-
Resource Management - Container Service
-
Resource Management - Data Migration
-
Resource Management - Durable Task
-
Resource Management - Key Vault
-
Resource Management - Quota
-
Resource Management - Sitemanager
-
Resource Management - Storage
-
Resource Management - Storage Cache
-
Resource Management - Storagediscovery
-
unknown
Patch Updates (9)
-
AI Evaluation
-
Identity
-
Resource Management - Compute
-
Resource Management - Data Protection
-
Resource Management - Key Vault
-
Resource Management - Quota
-
Resource Management - Sitemanager
-
Resource Management - Storage Cache
-
Resource Management - Storagediscovery
Beta Packages (7)
-
AI Agents
-
OpenTelemetry Exporter
-
Resource Management - Disconnectedoperations
-
Resource Management - Mongo Cluster
-
Resource Management - NetApp Files
-
Resource Management - SQL
-
unknown
Release highlights
AI Agents 1.2.0b5 Changelog
Features Added
- Added
run_handler
parameter toruns.create_and_process
allowing to make function tool calls manually or approve mcp tool calls.
Bugs Fixed
- Fixed regression, reverted ToolOutput type signature and usage in tool_output submission.
- Added
RunStepDeltaComputerUseDetails
andRunStepDeltaComputerUseToolCall
classes for streaming computer use scenarios. - Added
RunStepDeltaChunk
toStreamEventData
model (GitHub issues 43022)
AI Evaluation 1.12.0 Changelog
Features Added
- AOAI Graders now accept a “credential” parameter that can be used for authentication with an AzureOpenAIModelConfiguration
- Added
is_reasoning_model
parameter support toCoherenceEvaluator
,FluencyEvaluator
,SimilarityEvaluator
,GroundednessEvaluator
,RetrievalEvaluator
, andRelevanceEvaluator
to enable reasoning model configuration for o1/o3 models.
Bugs Fixed
- Support for multi-level nesting in OpenAI grader (experimental)
AI Evaluation 1.11.2 Changelog
Bugs Fixed
- **kwargs in an evaluator signature receives input columns that are not otherwise named in the evaluator’s signature
Azure AI Search 11.6.0 Changelog
Breaking Changes
VectorSearchCompression.rerankWithOriginalVectors
andVectorSearchCompression.defaultOversampling
don’t work with2025-09-01
and were replaced byVectorSearchCompression.RescoringOptions.enabledRescoring
andVectorSearchCompression.RescoringOptions.defaultOversampling
. If using2024-07-01
continue using the old properties, otherwise if using2025-09-01
use the new properties inRescoringOptions
.
Features Added
- Added
azure.search.documents.DocumentDebugInfo
. - Added
azure.search.documents.QueryDebugMode
. - Added
azure.search.documents.QueryResultDocumentSubscores
. - Added
azure.search.documents.SingleVectorFieldResult
. - Added
azure.search.documents.TextResult
. - Added
azure.search.documents.VectorsDebugInfo
. - Added new parameter
debug
inazure.search.documents.SearchClient.search
. - Added
azure.search.documents.indexes.LexicalNormalizer
. - Added
azure.search.documents.indexes.LexicalNormalizerName
. - Added
azure.search.documents.indexes.AnalyzeTextOptions.normalizer_name
. - Added
azure.search.documents.indexes.CustomNormalizer
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkill
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkillExtractionOptions
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkillChunkingProperties
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkillChunkingUnit
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkillMarkdownHeaderDepth
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkillOutputFormat
. - Added
azure.search.documents.indexes.DocumentIntelligenceLayoutSkillOutputMode
. - Added
azure.search.documents.indexes.RankingOrder
. - Added
azure.search.documents.indexes.RescoringOptions
. - Added
azure.search.documents.indexes.SearchField.normalizer_name
. - Added
azure.search.documents.indexes.SearchIndex.normalizer
. - Added
azure.search.documents.indexes.SearchIndexerKnowledgeStoreParameters
. - Added
azure.search.documents.indexes.VectorSearchCompressionRescoreStorageMethod
. - Support for running
VectorQuery
s against sub-fields of complex fields. - Added support for
2025-09-01
service version. - Support for reranker boosted scores in search results and the ability to sort results on either reranker or reranker
boosted scores in
SemanticConfiguration.rankingOrder
. - Support for
VectorSearchCompression.RescoringOptions
to configure how vector compression handles the original vector when indexing and how vectors are used during rescoring. - Added
SearchIndex.description
to provide a textual description of the index. - Support for
LexicalNormalizer
when definingSearchIndex
,SimpleField
, andSearchableField
and the ability to use it when analyzing text withSearchIndexClient.analyzeText
andSearchIndexAsyncClient.analyzeText
. - Support
DocumentIntelligenceLayoutSkill
skillset skill andOneLake
SearchIndexerDataSourceConnection
data source. - Support for
QueryDebugMode
in searching to retrieve detailed information about search processing. Onlyvector
is supported forQueryDebugMode
.
Identity 1.25.1 Changelog
Other Changes
- When
AZURE_TOKEN_CREDENTIALS
is set toManagedIdentityCredential
,DefaultAzureCredential
now skips the IMDS endpoint probe request and directly attempts token acquisition with full retry logic, matching the behavior of usingManagedIdentityCredential
standalone. (#43080) - Improved error messages from
ManagedIdentityCredential
to include the full error response from managed identity endpoints for better troubleshooting. (#43231)
OpenTelemetry Exporter 1.0.0b43 Changelog
Features Added
- OneSettings control plane: Add live metrics feature control (#43032)
- Adding customer sdkstats to feature statsbeat (#43066)
- OneSettings control plane: Add killswitch + exponential backoff + sdkstats feature control (#43147)
- Performance Counters (#43024)
- Performance Counters: Include Consumer spans (#43261)
Bugs Fixed
- Fixes issues #42897 - TypeError in _transmit_from_storage: LocalFileBlob.get() returns None (#42897)
- Fix logic for message field on Microsoft.ApplicationInsights.Message to sync with Breeze (#43060)
- Skip tests from test_storage.py causing transient build pipeline failures (#43122)
- Fix spelling for OTEL_PYTHON_EXCLUDED_URLS in samples (#43141)
- Revert #43060 (#43087)
Resource Management - Cognitive Services 14.0.0 Changelog
Resource Management - Compute 37.0.0 Changelog
Resource Management - Compute 37.0.1 Changelog
Resource Management - Container Service 40.0.0 Changelog
Resource Management - Data Migration 10.1.0 Changelog
Resource Management - Data Protection 2.0.1 Changelog
Resource Management - Disconnectedoperations 1.0.0b2 Changelog
Resource Management - Durable Task 1.0.0 Changelog
Resource Management - Key Vault 12.1.0 Changelog
Resource Management - Key Vault 12.1.1 Changelog
Resource Management - Mongo Cluster 1.1.0b3 Changelog
Resource Management - NetApp Files 14.0.0b2 Changelog
Resource Management - NetApp Files 14.0.0b3 Changelog
Resource Management - Quota 3.0.0 Changelog
Resource Management - Quota 3.0.1 Changelog
Resource Management - SQL 4.0.0b24 Changelog
Resource Management - Sitemanager 1.0.0 Changelog
Resource Management - Sitemanager 1.0.1 Changelog
Resource Management - Storage 23.1.0 Changelog
Resource Management - Storage Cache 3.0.0 Changelog
Resource Management - Storage Cache 3.0.1 Changelog
Resource Management - Storagediscovery 1.0.0 Changelog
Resource Management - Storagediscovery 1.0.1 Changelog
unknown 1.0.0b5 Changelog
Breaking Changes
- Cross-Language Package Identity Update: Updated package ID from
VoiceLive
toVoiceLive.WebSocket
for better cross-language consistency - Model Refactoring:
- Renamed
UserContentPart
toMessageContentPart
for clearer content part hierarchy - All message items now require a
content
field with list ofMessageContentPart
objects OutputTextContentPart
now inherits fromMessageContentPart
instead of being standalone- Enhanced Type Safety:
- Azure voice classes now use
AzureVoiceType
enum discriminators instead of string literals - Message role discriminators now use
MessageRole
enum values for better type safety - Removed Deprecated Parameters: Completely removed deprecated parameters from semantic detection classes:
- Removed
threshold
parameter from all semantic detection classes (AzureSemanticDetection
,AzureSemanticDetectionEn
,AzureSemanticDetectionMultilingual
) - Removed
timeout
parameter from all semantic detection classes - Users must now use
threshold_level
andtimeout_ms
parameters respectively - Removed Synchronous API: Completely removed synchronous WebSocket operations to focus exclusively on async patterns:
- Removed sync
connect()
function and syncVoiceLiveConnection
class from main patch implementation - Removed sync
basic_voice_assistant.py
sample (only async version remains) - Simplified sync patch to minimal structure with empty exports
- All functionality now available only through async patterns
- Updated Dependencies: Modified package dependencies to reflect async-only architecture:
- Moved
aiohttp>=3.9.0,<4.0.0
from optional to required dependency - Removed
websockets
optional dependency as sync API no longer exists - Removed optional dependency groups
websockets
,aiohttp
, andall-websockets
- Model Rename:
- Renamed
AudioInputTranscriptionSettings
toAudioInputTranscriptionOptions
for consistency with naming conventions - Renamed
AzureMultilingualSemanticVad
toAzureSemanticVadMultilingual
for naming consistency with other multilingual variants - Enhanced Type Safety: Turn detection discriminator types now use enum values instead of string literals for better type safety
Features Added
- Enhanced Semantic Detection Type Safety: Added new
EouThresholdLevel
enum for better type safety in end-of-utterance detection: LOW
for low sensitivity threshold levelMEDIUM
for medium sensitivity threshold levelHIGH
for high sensitivity threshold levelDEFAULT
for default sensitivity threshold level- Improved Semantic Detection Configuration: Enhanced semantic detection classes with better type annotations:
threshold_level
parameter now supports both string values andEouThresholdLevel
enum- Cleaner type definitions for
AzureSemanticDetection
,AzureSemanticDetectionEn
, andAzureSemanticDetectionMultilingual
- Improved documentation for threshold level parameters
- Comprehensive Unit Test Suite: Added extensive unit test coverage with 200+ test cases covering:
- All enum types and their functionality
- Model creation, validation, and serialization
- Async connection functionality with proper mocking
- Client event handling and workflows
- Voice configuration across all supported types
- Message handling with content part hierarchy
- Integration scenarios and real-world usage patterns
- Recent changes validation and backwards compatibility
- API Version Update: Updated to API version
2025-10-01
(from2025-05-01-preview
) - Enhanced Type Safety: Added new
AzureVoiceType
enum with values for better Azure voice type categorization: AZURE_CUSTOM
for custom voice configurationsAZURE_STANDARD
for standard voice configurationsAZURE_PERSONAL
for personal voice configurations- Improved Message Handling: Added
MessageRole
enum for better role type safety in message items - Enhanced Model Documentation: Comprehensive documentation improvements across all models:
- Added detailed docstrings for model classes and their parameters
- Enhanced enum value documentation with descriptions
- Improved type annotations and parameter descriptions
- Enhanced Semantic Detection: Added improved configuration options for all semantic detection classes:
- Added
threshold_level
parameter with options:"low"
,"medium"
,"high"
,"default"
(recommended over deprecatedthreshold
) - Added
timeout_ms
parameter for timeout configuration in milliseconds (recommended over deprecatedtimeout
) - Video Background Support: Added new
Background
model for video background customization: - Support for solid color backgrounds in hex format (e.g.,
#00FF00FF
) - Support for image URL backgrounds
- Mutually exclusive color and image URL options
- Enhanced Video Parameters: Extended
VideoParams
model with: background
parameter for configuring video backgrounds using the newBackground
modelgop_size
parameter for Group of Pictures (GOP) size control, affecting compression efficiency and seeking performance- Improved Type Safety: Added
TurnDetectionType
enum for better type safety and IntelliSense support - Package Structure Modernization: Simplified package initialization with namespace package support
- Enhanced Error Handling: Added
ConnectionError
andConnectionClosed
exception classes to the async API for better WebSocket error management
unknown 1.0.0 Changelog
Breaking Changes
- Improved Naming Conventions: Updated model and enum names for better clarity and consistency:
OAIVoice
enum renamed toOpenAIVoiceName
for more descriptive namingToolChoiceObject
model renamed toToolChoiceSelection
for better semantic meaningToolChoiceFunctionObject
model renamed toToolChoiceFunctionSelection
for consistency- Updated type unions and imports to reflect the new naming conventions
- Cross-language package mappings updated to maintain compatibility across SDKs
- Session Model Architecture: Separated
ResponseSession
andRequestSession
models for better design clarity: ResponseSession
no longer inherits fromRequestSession
and now inherits directly from_Model
- All session configuration fields are now explicitly defined in
ResponseSession
instead of being inherited - This provides clearer separation of concerns between request and response session configurations
- May affect type checking and code that relied on the previous inheritance relationship
- Model Cleanup: Removed unused
AgentConfig
model and related fields from the public API: AgentConfig
class has been completely removed from imports and exportsagent
field removed fromResponseSession
model (including constructor parameter)- Updated cross-language package mappings to reflect the removal
- Model Naming Convention Update: Renamed
EOUDetection
toEouDetection
for better naming consistency: - Class name changed from
EOUDetection
toEouDetection
- All inheritance relationships updated:
AzureSemanticDetection
,AzureSemanticDetectionEn
, andAzureSemanticDetectionMultilingual
now inherit fromEouDetection
- Type annotations updated in
AzureSemanticVad
,AzureSemanticVadEn
,AzureSemanticVadMultilingual
, andServerVad
classes - Import statements and exports updated to reflect the new naming
- Enhanced Content Part Type Safety: Content part discriminators now use enum values instead of string literals:
InputAudioContentPart.type
now usesContentPartType.INPUT_AUDIO
instead of"input_audio"
InputTextContentPart.type
now usesContentPartType.INPUT_TEXT
instead of"input_text"
OutputTextContentPart.type
now usesContentPartType.TEXT
instead of"text"
Features Added
- Enhanced WebSocket Connection Options: Significantly improved WebSocket connection configuration with transport-agnostic design:
- Added new timeout configuration options:
receive_timeout
,close_timeout
, andhandshake_timeout
for fine-grained control - Enhanced
compression
parameter to support both boolean and integer types for advanced zlib window configuration - Added
vendor_options
parameter for implementation-specific options passthrough (escape hatch for advanced users) - Improved documentation with clearer descriptions for all connection parameters
- Better support for common aliases from other WebSocket ecosystems (
max_size
,ping_interval
, etc.) - More robust option mapping with proper type conversion and safety checks
- Enhanced Type Safety: Improved type safety for content parts with proper enum usage:
InputAudioContentPart
,InputTextContentPart
, andOutputTextContentPart
now useContentPartType
enum values instead of string literals- Better IntelliSense support and compile-time type checking for content part discriminators
unknown 1.1.0b1 Changelog
Features Added
- AgentConfig Support: Re-introduced
AgentConfig
functionality with enhanced capabilities: AgentConfig
model added back to public API with full import and export supportagent
field re-added toResponseSession
model for session-level agent configuration- Updated cross-language package mappings to include
AgentConfig
support - Provides foundation for advanced agent configuration scenarios
Latest Releases
View all the latest versions of Python packages here.
Installation Instructions
To install the latest beta version of the packages, copy and paste the following commands into a terminal:
$> pip install azure-ai-agents==1.2.0b5
$> pip install azure-ai-evaluation==1.12.0
$> pip install azure-ai-evaluation==1.11.2
$> pip install azure-ai-voicelive==1.0.0b5
$> pip install azure-ai-voicelive==1.0.0
$> pip install azure-ai-voicelive==1.1.0b1
$> pip install azure-identity==1.25.1
$> pip install azure-mgmt-cognitiveservices==14.0.0
$> pip install azure-mgmt-compute==37.0.0
$> pip install azure-mgmt-compute==37.0.1
$> pip install azure-mgmt-containerservice==40.0.0
$> pip install azure-mgmt-datamigration==10.1.0
$> pip install azure-mgmt-dataprotection==2.0.1
$> pip install azure-mgmt-disconnectedoperations==1.0.0b2
$> pip install azure-mgmt-durabletask==1.0.0
$> pip install azure-mgmt-keyvault==12.1.0
$> pip install azure-mgmt-keyvault==12.1.1
$> pip install azure-mgmt-mongocluster==1.1.0b3
$> pip install azure-mgmt-netapp==14.0.0b2
$> pip install azure-mgmt-netapp==14.0.0b3
$> pip install azure-mgmt-quota==3.0.0
$> pip install azure-mgmt-quota==3.0.1
$> pip install azure-mgmt-sitemanager==1.0.0
$> pip install azure-mgmt-sitemanager==1.0.1
$> pip install azure-mgmt-sql==4.0.0b24
$> pip install azure-mgmt-storage==23.1.0
$> pip install azure-mgmt-storagecache==3.0.0
$> pip install azure-mgmt-storagecache==3.0.1
$> pip install azure-mgmt-storagediscovery==1.0.0
$> pip install azure-mgmt-storagediscovery==1.0.1
$> pip install azure-monitor-opentelemetry-exporter==1.0.0b43
$> pip install azure-search-documents==11.6.0
Feedback
If you have a bug or feature request for one of the libraries, please post an issue to GitHub.