Skip to content

CHANGELOG.md

Latest commit

 

History

History
977 lines (555 loc) · 63.8 KB

CHANGELOG.md

File metadata and controls

977 lines (555 loc) · 63.8 KB

Changelog

1.52.0 (2026-04-30)

Features

  • [Interactions] Add FileCitation.{custom_metadata,media_id,page_number} (a09716e)
  • Add output_info to BatchJob (e337ba9)
  • Add ImageResizeMode for GenerateVideos (e089fcb)
  • Add new Gemini Deep Research agent models (88d7019)
  • Add Vertex Dataset input and output options for batch jobs (612601d)
  • interaction-api: Add grounding tool usage breakdown to Interaction Usage. (bc2f815)
  • introduce enterprise and GOOGLE_GENAI_USE_ENTERPRISE (97adcd9)
  • Replace the more ambiguous rate field with sample_rate. (d762b6e)

Documentation

  • replace Vertex AI with Gemini Enterprise Agent Platform (84c1d43)
  • update doc string to replace Vertex AI with Gemini Enterprise Agent Platform, update method error message to replace Vertex AI with Gemini Enterprise Agent Platform (previously known as Vertex AI), update converter error message to replace Vertex AI with Gemini Enterprise Agent Platform (c5aefa7)
  • update Gemini Enterprise Agent Platform home page url (7b46a6f)

1.51.0 (2026-04-16)

Features

  • Add gemini-3.1-flash-tts-preview model to options (d9a5733)

Bug Fixes

  • ResponseStream fails to parse error message and multi-line SSE data payloads (3a1cb22)

1.50.0 (2026-04-14)

Features

  • [Experimental] Add interactions support. (abc1ee4)
  • Add "eu" as a supported service location for Vertex AI platform. (78e1bc1)
  • Add Live Avatar new fields (8e8146a)
  • Add webhook_config to batches.create() and models.generate_videos() (ac0e49e)

Bug Fixes

  • fix broken javadoc by switching to dokka. (4b76baf)
  • internal workflow change (b80b142)

1.49.0 (2026-04-14)

Bug Fixes

  • fix broken javadoc by switching to dokka. (4b76baf)
  • internal workflow change (b80b142)

1.48.0 (2026-04-13)

Features

  • [Experimental] Add interactions support. (abc1ee4)
  • Add "eu" as a supported service location for Vertex AI platform. (78e1bc1)
  • Add Live Avatar new fields (8e8146a)
  • Add webhook_config to batches.create() and models.generate_videos() (ac0e49e)

1.47.0 (2026-04-08)

Documentation

  • Remove deprecated product recontext model samples from docstrings (6f73c4a)

1.46.0 (2026-04-01)

Bug Fixes

  • Fix service_tier enums. (d4641e4)

1.45.0 (2026-03-28)

Features

  • Add consent_audio and voice_consent_signature and AsyncSession.setup_complete (959c01b)
  • Add labels field to Veo configs (86f235a)
  • Add model_status to GenerateContentResponse (Gemini API only) (6c35249)
  • Add part_metadata in Part (Gemini API only) (6c35249)
  • Add service tier to GenerateContent. (5f61a16)
  • genai: add TURN_INCLUDES_AUDIO_ACTIVITY_AND_ALL_VIDEO to TurnCoverage (dbc4c33)
  • support hyperparameters in distillation tuning (fdedc74)
  • Support rendered_parts in GroundingSupport (6c35249)

Bug Fixes

  • support us region routing (3296bfe)

1.44.0 (2026-03-18)

Features

  • Support include_server_side_tool_invocations for genai. (d92fdb1)

1.43.0 (2026-03-12)

Features

  • Add inference_generation_config to EvaluationConfig for Tuning (c9632f0)
  • Allow custom endpoints for authentication with Vertex AI in Java (5b38728)
  • enable language code for audio transcription config in Live API for Vertex AI (fb034ff)

1.42.0 (2026-03-04)

Features

  • Update data types from discovery doc. (850d527)

Bug Fixes

  • initialize comprehensive and curated history in constructor (f74a426)

1.41.0 (2026-02-26)

Features

  • Add Image Grounding support to GoogleSearch tool (0daefbc)
  • enable server side MCP and disable all other AFC when server side MCP is configured. (498a2c4)
  • examples (917aee0)
  • initial integration (917aee0)
  • set up pom + fix test (917aee0)

Bug Fixes

  • use NonClosingDelegatingHttpClient (917aee0)

Documentation

  • explain how to run tests (4f6a811)

1.40.0 (2026-02-19)

Features

  • Add registerFiles for Java. (ce0b638)
  • Add UnifiedMetric support to Vertex Tuning evaluation config (a28ebfc)
  • Support multimodal embedding for Gemini Embedding 2.0 and support MaaS models in Models.embed_content() (Vertex AI API) (23a7913)

1.39.0 (2026-02-07)

Features

  • Support encryption_spec in tuning job creation configuration for GenAI SDK (7a4fb51)

1.38.0 (2026-02-05)

Features

Bug Fixes

  • 'No SLF4J providers were found' on examples (97fa11f)
  • Make Apache HTTP Components an optional Maven dependency in GenAI. (a01e464)
  • Replace System .err & .out with correct (JUL) Logging in GenAI AsyncLive. (1f756d6)

1.37.0 (2026-01-30)

Features

  • Support distillation tuning (cec1b88)
  • Support OSS Tuning in GenAI SDK (868d8ed)

Bug Fixes

  • Add metadata in batch inlined response (c80dd07)

1.36.0 (2026-01-22)

Features

  • Add ModelArmorConfig support for prompt and response sanitization via the Model Armor service (9c77a8f)

1.35.0 (2026-01-14)

Features

1.34.0 (2026-01-08)

Features

  • Add gemini-3-pro-preview support for local tokenizer (40480f4)

Documentation

  • Update Virtual Try-On model id in samples and docstrings (e349635)

1.33.0 (2026-01-07)

Note: The artifacts for this version were not published to Maven. Please upgrade directly to version 1.34.0.

1.32.0 (2025-12-17)

Features

  • Add minimal and medium thinking levels. (ecfadfe)
  • Add PersonGeneration to ImageConfig for Vertex Gempix (d8a4c43)
  • Add ultra high resolution to the media resolution in Parts. (7c9b7f6)
  • support multi speaker for Vertex AI (c50e47b)

1.31.0 (2025-12-11)

Features

  • Add enableEnhancedCivicAnswers feature in GenerateContentConfig (0570478)
  • Add IMAGE_RECITATION and IMAGE_OTHER enum values to FinishReason (26c3c35)
  • Add voice activity detection signal. (d1ca685)

Documentation

  • Add an example for ClientOptions usage (0a8a26e)

1.30.0 (2025-12-08)

Features

  • Add ProxyOptions in ClientOptions for configuring proxies (eafdf79)
  • Ephemeral token for Gemini Live API support in Java (4ce094b)
  • Support ReplicatedVoiceConfig (dbe314d)

1.29.0 (2025-12-03)

Features

  • Add empty response for tunings.cancel() (57218f5)

Bug Fixes

  • Match the versions of the Java Protobuf and Google API Common dependencies (com.google.protobuf:protobuf-java:3.25.5 & com.google.api:api-common:2.45.0) with Google Cloud Java SDKs. (6c37f58)

Documentation

  • Recommend using response_json_schema in error messages and docstrings. (6b952e9)

1.28.0 (2025-11-17)

Features

  • add display name to FunctionResponseBlob (8db8c57)
  • add display name to FunctionResponseFileData (f5ee8b7)
  • Add generate_content_config.thinking_level (a47df92)
  • Add image output options to ImageConfig for Vertex (3eac0b8)
  • Add part.media_resolution (a47df92)
  • support Function call argument streaming for all languages (f310452)
  • support upload to file search stores (7862ce3)

1.27.0 (2025-11-12)

Features

  • Add images() convenience method to GenerateImagesResponse (155df8d)
  • Add EvaluationConfig support to tune() in Java (795cf73)
  • Automatically set response type in FunctionDeclaration during the AFC (5ce99df)
  • Support overriding the max read length in the JSON parser (29d2fca)

Bug Fixes

  • Add missing fields to the model types (7b7b41f)
  • Fix base_steps parameter for recontext_image (85aaa79)
  • Fix models.list() filter parameter (123ada5)

Documentation

  • Add README for Files API (6d206aa)

1.26.0 (2025-11-05)

Features

  • Add clearXxx methods to data type builders (a4900c9)
  • add complete stats to BatchJob (659c65c)
  • Add FileSearch tool and associated FileSearchStore management APIs (8ada6ef)
  • Add image_size to ImageConfig (Early Access Program) (c1af981)
  • Added phish filtering feature. (ed4e12c)
  • Return response headers for generateContentStream (82a8118)
  • Support lists as function parameters in AFC (fixes #527) (452d2e5)

Bug Fixes

  • disable AFC when there are AFC incompatible tool presented. (6099d87)

Documentation

  • add blank line before version update marker (82616c2)
  • Update Java SDK README spacing (82616c2)

1.25.0 (2025-10-29)

Features

  • Add safety_filter_level and person_generation for Imagen upscaling (09a8075)
  • Add support for preference optimization tuning in the SDK. (5d4123c)
  • Added Operations.get which is a generic method which will handle all Operation types. (c1dc32f)
  • Pass file name to the backend when uploading with a file path (081a9a6)
  • support default global location when not using api key with vertexai backend (f9028a7)
  • Support retries in API requests (3d5de00)

Documentation

  • Add docstring for classes and fields that are not supported in Gemini or Vertex API (7a03dac)
  • Add docstring for enum classes that are not supported in Gemini or Vertex API (830a12f)
  • Add documentation for the retry behavior (4fbcf51)

1.24.0 (2025-10-22)

Features

  • Add enable_enhanced_civic_answers in GenerationConfig (684a2c5)
  • support createEmbeddings in Batches.java (8947f6f)
  • support jailbreak in HarmCategory and BlockedReason (3dab40b)

Bug Fixes

  • Make async methods in Batches module truly non-blocking (f2ae75a)
  • Make async methods in Caches, Tuning, and Operations modules truly non-blocking (db56239)
  • Make async methods in Models module truly non-blocking (c205d01)

1.23.0 (2025-10-15)

Features

  • Support video extension for Veo on Gemini Developer API (b398509)

1.22.0 (2025-10-10)

Features

  • Enable Google Maps tool for Genai. (a4baf3c)
  • Support enableWidget feature in GoogleMaps (aefbd5c)
  • Support Gemini batch inline request's metadata and add test coverage to safety setting (17033b3)

1.21.0 (2025-10-08)

Features

  • Add NO_IMAGE enum value to FinishReason (6b00c0b)
  • Add labels field to Imagen configs (e69cf68)
  • Add thinking_config for live (274c21d)
  • Add utility methods for creating FunctionResponsePart and creating FunctionResponse Part with FunctionResponseParts (af16a4c)
  • Enable Ingredients to Video and Advanced Controls for Veo on Gemini Developer API (Early Access Program) (4c42e65)

Bug Fixes

  • Ensure Live server message are properly converted (206dc88)

1.20.0 (2025-10-01)

Features

  • Add ImageConfig to GenerateContentConfig (6fb5eba)

1.19.0 (2025-09-30)

Features

  • expose session id in Live API (b6d5389)
  • rename ComputerUse tool (early access) (4bbba2b)

1.18.0 (2025-09-25)

Features

  • Add FunctionResponsePart & ToolComputerUse.excludedPredefinedFunctions (1a24bed)
  • Support Imagen 4 Ingredients on Vertex (b5eed8d)

Bug Fixes

  • Expose JOB_STATE_RUNNING and JOB_STATE_EXPIRED for Gemini Batches states (c5b4fdf)
  • initialization of pre_tuned_model_checkpoint_id from tuning config. (c293633)
  • Make async generateContent and generateContentStream truly non-blocking (5cb18fd)
  • only run unit tests in github action (9b2861b)

1.17.0 (2025-09-16)

Features

  • Add 'turn_complete_reason' and 'waiting_for_input' fields. (5bc4873)
  • Add VideoGenerationMaskMode enum for Veo 2 Editing (e5c8277)
  • Add labels to create tuning job config (695e17a)
  • generate the function_call class's converters (38703c7)
  • java local tokenizer (d774185)
  • Support Veo 2 Editing on Vertex (d401d3c)

Bug Fixes

  • Enable id field in FunctionCall for Vertex AI. (3773fe7)
  • update Live API audio example with better interruption handling (cad8df9)

1.16.0 (2025-09-02)

Features

  • Add resolution field for Gemini Developer API Veo 3 generation (eec410c)
  • add the response body for generateContent (a011580)

Documentation

  • Refactor/update docstrings for Imagen and Veo (2470101)

1.15.0 (2025-08-27)

Features

  • add sdkHttpResponse.headers to *Delete responses. (4be038d)
  • Add output_gcs_uri to Imagen upscale_image (7649467)
  • add the response body for generateContent (6e28ab4)
  • add the response body for generateContent (b2a5b3f)
  • Add VALIDATED mode into FunctionCallingConfigMode (4bb8680)
  • Add VideoGenerationReferenceType enum for generate_videos (df9d910)
  • Support GenerateVideosSource for Veo GenerateVideos (c26af63)
  • support tunings.cancel in the genai SDK for Python, Java, JS, and Go (9982251)

Documentation

  • Refactor model IDs into a Constants class (dacd787)

1.14.0 (2025-08-22)

Features

  • Add add_watermark field for recontext_image (Virtual Try-On, Product Recontext) (5aacbc0)

Bug Fixes

  • Fix the bug that files.create doesn't return the upload URL correctly (eb40c5f)

Documentation

  • update TokensInfo docstring (48eba7f)

1.13.0 (2025-08-18)

Features

  • expose JsonSerializable.stringToJsonNode to help user better use *JsonSchema fields. (35d783b)
  • Return response headers for all methods (except streaming methods) (7e8b71b)
  • Support Imagen image segmentation on Vertex (e2a561b)
  • Support Veo 2 Reference Images to Video Generation on Vertex (2f5580f)

1.12.0 (2025-08-13)

Features

  • enable continuous fine-tuning on a pre-tuned model in the SDK. (e49d350)
  • support document name in grounding metadata (8273922)
  • Support exclude_domains in Google Search and Enterprise Web Search (e975d28)

1.11.0 (2025-08-06)

Features

  • Add image_size field for Gemini Developer API Imagen 4 generation (c50c755)
  • enable responseId for Gemini Developer API (4912ff4)
  • support extraBody in HttpOptions class (036bac8)
  • Support image recontext on Vertex (e7de8c8)
  • Support new enum types for UrlRetrievalStatus (cb27222)
  • support response headers in Go for all methods. (222b41e)

Bug Fixes

  • Remove duplicate JavaTimeModule in JsonSerializable (a7dbd4c)

Documentation

  • Add Imagen and Veo to README (cc0a0aa)
  • Add latest models features in README (a2eccaf)
  • mark Client as thread safe and Chat as not thread safe (be3e50e)

1.10.0 (2025-07-23)

Features

  • Add image_size field for Vertex Imagen 4 generation (950c0c6)
  • Support API keys for VertexAI mode for Java SDK (826c0dc)
  • Support http headers in GenerateContentResponse (5282774)

Bug Fixes

  • Defer loading ADC when credentials is provided explicitly in Live API (a540614)
  • live: Enhance security by moving api key from query parameters to header (e48c7f1)
  • Pager throws an exception if list request returns nothing(correct behavior is returning a Pager without any items in it) (0a2301b)

Documentation

  • Update README with latest features in Client (dcf70cc)

1.9.0 (2025-07-16)

Features

  • Add addWatermark parameter to the edit image configuration. (c4598da)
  • add Tuning support for Java (0ab209d)

Documentation

  • Update generated video resolution config docstring (9a2ced8)

1.8.0 (2025-07-09)

Features

  • Add new languages for Imagen 4 prompt language (7e1e6d2)

1.7.0 (2025-07-01)

Features

  • Support Batches delete (782465d)
  • Support different media input in Vertex Live API (7f4c6bf)

Bug Fixes

1.6.0 (2025-06-25)

Features

  • Add compressionQuality enum for generate_videos (b0e665b)
  • Add enhance_input_image and image_preservation_factor fields for upscale_image (94a329a)
  • allow users to access headers for generateContent method and generateContentStream (0315357)
  • Batches support in Java (5ce13e9)
  • configure release-please to automatically update package version across all files during releases. (9131ac2)
  • expose the responseJsonSchema in GenerateContentConfig (9d9acdb)
  • support client.caches.update method (345c2b9)

Documentation

  • add more comments to make it easier to follow live api code, and to explain the usage of new concepts like thenCompose. (96c792c)
  • improve generate images documentation (44c21dd)
  • Update description of thinking_budget. (265f20a)

1.5.0 (2025-06-19)

Features

  • Add a default 5 minutes timeout to the Java SDK. (d80d23b)
  • enable json schema for controlled output and function declaration. (35d93d8)
  • Parametrize model ID in Java genai samples and set hardcoded model IDs as default values if model ID not passed in by user. (c1b0948)
  • resolve deep conditional nesting in Live API examples for better readability. (aa71865)

Documentation

  • fix small typo in comments (a6bfe36)

1.4.1 (2025-06-12)

Bug Fixes

  • Add backwards compatibility for generateVideos in Java SDK (504984a)

Miscellaneous Chores

1.4.0 (2025-06-11)

Features

  • Add datastore_spec field for VertexAISearch (8c8b47a)
  • Add support for Veo frame interpolation and video extension (94f5e41)
  • RAG - Introducing context storing for Gemini Live API. (d63d625)
  • Support maxConnections and maxConnectionsPerHost in Client instantiation (7a4a8ac)
  • Support passing builder classes to setter methods (996c994)
  • Support passing builder varargs to setter methods (0b08524)
  • Support varargs in Enum setter methods (cd63fd0)
  • Support varargs in setter methods (0b08524)

Documentation

  • Add javadoc for setter methods in types classes (996c994)

1.3.0 (2025-06-04)

Features

  • Add enhance_prompt field for Gemini Developer API generate_videos (944af16)
  • Enable url_context for Vertex (cc772d8)
  • java: Support GEMINI_API_KEY as environment variable for setting API key. (84e7588)

Documentation

  • Updating docs to include how to initialize client via environment variables (7fc3e62)

1.2.0 (2025-05-30)

Features

  • Adding thought_signature field to the Part to store the signature for thoughts. (dc9c8f7)
  • include UNEXPECTED_TOOL_CALL enum value to FinishReason for Vertex AI APIs. (8186ff0)

Bug Fixes

  • Rename LiveEphemeralParameters to LiveConnectConstraints. (0655609)

1.1.0 (2025-05-28)

Features

  • Add generate_audio field for private testing of video generation (16f2b0f)
  • support new fields in FileData, GenerationConfig, GroundingChunkRetrievedContext, RetrievalConfig, Schema, TuningJob, VertexAISearch, (6bf3d32)

Documentation

  • Fix comment typo for Modality.AUDIO (18dea48), closes #620

1.0.0 (2025-05-19)

⚠ BREAKING CHANGES

  • Support java.time.Instant and java.time.Duration for time/duration fields

Features

  • add time range filter to Google Search Tool (4aba9a6)
  • Add async list methods to Models, Caches, and Files (cde10ab)
  • add async streaming chat (df6446a)
  • add async support for chat (9973df7)
  • Add AsyncPager class (cde10ab)
  • Add basic support for async function calling. (653094a)
  • add Caches.create/delete/get support (c4d17a0)
  • Add client.models.list() method (25883eb)
  • Add Files module with Files.upload, .get and .delete (e211ca1)
  • Add Files.download methods (082c890)
  • Add Image.fromFile() support in Java SDK (4ddcaa1)
  • Add list methods in Caches and Files module (ce6ed85)
  • add live proactivity_audio and enable_affective_dialog (64c20c7)
  • add multi-speaker voice config (1689f77)
  • Add pagination support (25883eb)
  • Add support for lat/long in search. (f43f00f)
  • Add Video FPS, and enable start/end_offset for MLDev (bc1df02)
  • Added Async wrappers for the Files Module operations (b5b608d)
  • enable automatic function calling for Models.generateContent method (ecca800)
  • record automatic function calling history for Chat.sendMessage methods. (7428d87)
  • support customer-managed encryption key in cached content (3021aaf)
  • Support ephemeral token creation in Python (7f33c20)
  • Support java.time.Instant and java.time.Duration for time/duration fields (460d3d3)
  • Support models.get/delete/update in Java (3d8d8ec)
  • Support Url Context Retrieval tool (f4fd37a)
  • Support using ephemeral token in Live session connection in Python (7f33c20)

Bug Fixes

  • bug in validate history (7428d87)
  • Transformers are not being used in list types (a9609db)

Miscellaneous Chores

0.8.0 (2025-05-13)

⚠ BREAKING CHANGES

  • mark *Parameters classes as internal
  • add enum support

Features

  • add Content.text() quick accessor (c184615)
  • add enum support (86b0553)
  • Add httpOptions field to type classes (971177d)
  • add support for audio, video, text and session resumption. (dcd08a4)
  • automatically parse a java.lang.reflect.Method instance into a FunctionDeclaration when users pass it in as a Tool. (e5109a0)
  • enable request level http options(set through XxxConfig) (abbe820)
  • support display_name for Blob class when calling Vertex AI (db33f8a)

Bug Fixes

  • fix import in live sample (78cc345)

Documentation

  • Improve docs for response_mime_type and response_schema. Relate to #297 (4782f56)

Miscellaneous Chores

  • mark *Parameters classes as internal (77c9d71)

0.7.0 (2025-05-06)

⚠ BREAKING CHANGES

  • Rename getVideoOperation to getVideosOperation for Java SDK

Features

  • Add Tool.enterprise_web_search field (919c22a)
  • Add computeTokens support (85c751e)
  • Add countTokens method support (8b8fb45)
  • Add support for Grounding with Google Maps (24b1127)
  • enable input transcription for Gemini API. (9065970)
  • http: Make HttpApiClient public (5da0746)

Documentation

  • Add javadoc to public classes/methods (78c2a39)

Code Refactoring

  • Rename getVideoOperation to getVideosOperation for Java SDK (4cd355b)

0.6.1 (2025-05-01)

Bug Fixes

  • generateContentStream failed to find converter method GenerateContentResponseFromXxx (590cb98)

Miscellaneous Chores

0.6.0 (2025-04-30)

⚠ BREAKING CHANGES

  • add streaming support for Chat

Features

  • add streaming support for Chat (59eca52)
  • add support for live grounding metadata (e90dd15)
  • Enable configurable speech detection in live SDK (d6bfa76)
  • introduce FunctionDeclaration.fromMethod class method to parse a FunctionDeclaration instance from a java.lang.reflect.Method instance. (aeb63b4)
  • make min_property, max_property, min_length, max_length, example, patter fields available for Schema class when calling Gemini API (8527db5)
  • Support continuous window compression in the Live API (7acb98c)
  • Support setting the default base URL in clients via Client.setDefaultBaseUrls() (c119778)

Bug Fixes

  • do not raise error for default field in Schema for Gemini API calls (8bb1c64)
  • Throw IllegalArgumentException rather than general Error when an invalid parameter is passed to Gemini/Vertex backend (1bd4485)
  • Update error message when a service from an unsupported backend is called (4cd81ea)
  • Update Live API websocket base url in java SDK (8c97572)

Documentation

  • add a link for where to find the Google Cloud project id, API key and location (3ac726c)

0.5.0 (2025-04-22)

Features

  • Populate X-Server-Timeout header when a request timeout is set. (7126995)
  • support default field in Schema when users call Gemini API (1d7c48d)

0.4.0 (2025-04-15)

⚠ BREAKING CHANGES

  • Hide internal classes ApiResponse and Common
  • Move generation config to the top level LiveConnectConfig
  • Remove AsyncSession.sendContent method which was accidentally added
  • No more checked exceptions
  • Remove AsyncSession.sendContent method which was accidentally added
  • No more checked exceptions

Features

  • Add async editImage support for Java SDK (358c896)
  • add Chat module (03442fc)
  • Add domain to Web GroundingChunk (990586b)
  • Add generationComplete notification to Live ServerContent (de5a762)
  • add session resumption. (5c35c52)
  • add support for model_selection_config to GenerateContentConfig (1958d4c)
  • Add thinking_budget to ThinkingConfig for Gemini Thinking Models (a3ea4ed)
  • Add traffic type to GenerateContentResponseUsageMetadata (fac241b)
  • Add types for configurable speech detection (021d74b)
  • Add types to support continuous sessions with a sliding window (d5e8a32)
  • Add UsageMetadata to LiveServerMessage (8b57e99)
  • Add Veo 2 generate_videos support in Java SDK (272649a)
  • expose generation_complete, input/output_transcription & input/output_audio_transcription to SDK for Vertex Live API (e3b517a)
  • merge GenerationConfig into LiveConnectConfig (e2fa59d)
  • No more checked exceptions (e452491)
  • No more checked exceptions (cad9eba)
  • Remove experimental warnings for generate_videos and operations (be93305)
  • Support audio transcription in Vertex Live API (1ddb6d4)
  • support media resolution (ce9e7b0)
  • Support RealtimeInputConfig, and language_code in SpeechConfig in python (f5bd874)
  • Update VertexRagStore (4a75670)

Bug Fixes

  • fix errors in live samples (5a9892f)
  • Move generation config to the top level LiveConnectConfig (c71e78e)
  • Remove AsyncSession.sendContent method which was accidentally added (e452491)
  • Remove AsyncSession.sendContent method which was accidentally added (cad9eba)
  • schema handling in transformer (8064028)
  • Use orElseGet to defer loading of default credentials. (bc19d79)
  • Use orElseGet to defer loading of default credentials. (#124) (6adac0e)

Documentation

  • Add examples for embedContent (7583ddf)
  • docstring improvements (9d0ca34)
  • Update examples and README after removing checked exceptions (5cb6ee5)

Code Refactoring

  • Hide internal classes ApiResponse and Common (6218143)

0.3.0 (2025-03-28)

⚠ BREAKING CHANGES

  • Only expose toJson and fromJson as public methods in JsonSerializable
  • Rename Client.isVertexAI() to Client.vertexAI() to align with other places
  • change bytes datatype from String to byte[] to avoid potential data loss

Features

  • Add client.models.embedContent methods (02f177c)
  • Add engine to VertexAISearch (8ed56ad)
  • add IMAGE_SAFTY enum value to FinishReason (02ee351)
  • Add Imagen edit_image support in Java SDK (4ae1366)
  • Add quick constructors for Content and Part (2ff9e86)
  • Add response_id and create_time to GenerateContentResponse (f52e068)
  • Add sendClientContent, sendRealtimeInput, sendToolResponse to live session (dbd9cf7)
  • add types for Live API (2cdb435)
  • allow title property to be sent to Gemini API. Gemini API now supports the title property, so it's ok to pass this onto both Vertex and Gemini API. (9afeedc)
  • enable minItem, maxItem, nullable for Schema type when calling Gemini API. (5975a55)
  • enable union type for Schema when calling Gemini API. (3d26960)
  • Save prompt safety attributes in dedicated field for generate_images (c8c5aaf)
  • Support aspect ratio for edit_image (7742349)
  • Support global endpoint in java natively (77e364e)
  • Support Live API (5741147)
  • support new UsageMetadata fields (062cbe0)
  • Support returned safety attributes for generate_images (ae6f5e4)
  • throw error when given method is not supported in Gemini API or Vertex AI (f055ad8)

Bug Fixes

  • change bytes datatype from String to byte[] to avoid potential data loss (9979823)
  • Getter methods in Client will return null rather than throw exceptions when the value is not set (c1b0290)
  • log warning instead of throw error in GenerateContentResponse.text() quick accessor when there are mixed types of parts in the response. (13c5ccf)
  • Only expose toJson and fromJson as public methods in JsonSerializable (dbef8b4)
  • Remove unsupported parameter negative_prompt from Gemini API generate_images (1d881d9)
  • Rename Client.isVertexAI() to Client.vertexAI() to align with other places (4211ca1)
  • schema transformer logic fix. (e6ab7d5)
  • set default scope in VertexAI client ADC (15aeede)
  • Set partial httpOptions in the Client will override all other options to empty (46369a7)

0.2.0 (2025-02-24)

Features

  • Add Imagen upscale_image support for Java (7eb172f)
  • Enable Media resolution for Gemini API. (30c8aa6)
  • provide executable code and code execution result quick accessors for GenerateContentResponse class (548f8c3)
  • support property_ordering in response_schema (fixes #236) (814aa6f)

Bug Fixes

  • Fix private class/method generation for Java SDK (0c4f1f8)
  • Set request content-type encoding to UTF-8 (#56) (09329f7)

Documentation

  • Update the model to gemini-2.0-flash-001 in the docs and the examples. (a4bc69a)