Skip to content

add changes to match the new api spec, add imagen, and google-ai support #6907

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 10 commits into from
Apr 25, 2025
Merged

add changes to match the new api spec, add imagen, and google-ai support #6907

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
f2dc583
add changes to match the new api spec, add imagen, and google-ai support
Apr 23, 2025
1c56f97
revert changes in vertexai dir
Apr 24, 2025
37d2b70
remerge main
Apr 24, 2025
1a22821
revert unintended vertex changes
Apr 24, 2025
274efa9
fixes for comments
Apr 25, 2025
22fd407
format pass
Apr 25, 2025
9c3455a
fix github workflow and failing test
Apr 25, 2025
4b9bd3a
format again
Apr 25, 2025
6c28594
fixed up docs
Apr 25, 2025
8ea196e
fixes for comments
Apr 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
97 changes: 0 additions & 97 deletions firebase-ai/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,100 +1,3 @@
# Unreleased
* [changed] **Breaking Change**: `LiveModelFutures.connect` now returns `ListenableFuture<LiveSessionFutures>` instead of `ListenableFuture<LiveSession>`.
* **Action Required:** Remove any transformations from LiveSession object to LiveSessionFutures object.
* **Action Required:** Change type of variable handling `LiveModelFutures.connect` to `ListenableFuture<LiveSessionsFutures>`
* [changed] **Breaking Change**: Removed `UNSPECIFIED` value for enum class `ResponseModality`
* **Action Required:** Remove all references to `ResponseModality.UNSPECIFIED`
* [changed] **Breaking Change**: Renamed `LiveGenerationConfig.setResponseModalities` to `LiveGenerationConfig.setResponseModality`
* **Action Required:** Replace all references of `LiveGenerationConfig.setResponseModalities` with `LiveGenerationConfig.setResponseModality`
* [feature] Added support for `HarmBlockThreshold.OFF`. See the
[model documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/configure-safety-filters#how_to_configure_content_filters){: .external}
for more information.
* [fixed] Improved thread usage when using a `LiveGenerativeModel`. (#6870)
* [fixed] Fixed an issue with `LiveContentResponse` audio data not being present when the model was
interrupted or the turn completed. (#6870)
* [fixed] Fixed an issue with `LiveSession` not converting exceptions to `FirebaseVertexAIException`. (#6870)


# 16.3.0
* [feature] Emits a warning when attempting to use an incompatible model with
`GenerativeModel` or `ImagenModel`.
* [changed] Added new exception type for quota exceeded scenarios.
* [feature] `CountTokenRequest` now includes `GenerationConfig` from the model.
* [feature] **Public Preview:** Added support for streaming input and output (including audio) using the [Gemini Live API](/docs/vertex-ai/live-api?platform=android)
**Note**: This feature is in Public Preview, which means that it is not subject to any SLA or deprecation policy and could change in backwards-incompatible ways.
* [changed] **Breaking Change**: `ImagenInlineImage.data` now returns the raw
image bytes (in JPEG or PNG format, as specified in
`ImagenInlineImage.mimeType`) instead of Base64-encoded data. (#6800)
* **Action Required:** Remove any Base64 decoding from your
`ImagenInlineImage.data` usage.
* The `asBitmap()` helper method is unaffected and requires no code changes.

# 16.2.0
* [fixed] Added support for new values sent by the server for `FinishReason` and `BlockReason`.
* [changed] Added support for modality-based token count. (#6658)
* [feature] Added support for generating images with Imagen models.

# 16.1.0
* [changed] Internal improvements to correctly handle empty model responses.

# 16.0.2
* [fixed] Improved error message when using an invalid location. (#6428)
* [fixed] Fixed issue where Firebase App Check error tokens were unintentionally missing from the requests. (#6409)
* [fixed] Clarified in the documentation that `Schema.integer` and `Schema.float` only provide hints to the model. (#6420)
* [fixed] Fixed issue were `Schema.double` set the format parameter in `Schema`. (#6432)

# 16.0.1
* [fixed] Fixed issue where authorization headers weren't correctly formatted and were ignored by the backend. (#6400)

# 16.0.0
* [feature] {{firebase_vertexai}} is now Generally Available (GA) and can be
used in production apps.

Use the {{firebase_vertexai_sdk}} to call the {{gemini_api_vertexai_long}}
directly from your app. This client SDK is built specifically for use with
Android apps, offering security options against unauthorized clients
as well as integrations with other Firebase services.

* If you're new to this library, visit the
[getting started guide](/docs/vertex-ai/get-started?platform=android).

* If you were using the preview version of the library, visit the
[migration guide](/docs/vertex-ai/migrate-to-ga?platform=android) to learn
about some important updates.
* [changed] **Breaking Change**: Changed `functionCallingConfig` parameter type to be nullable in `ToolConfig`. (#6373)
* [changed] **Breaking Change**: Removed `functionResponse` accessor method from `GenerateContentResponse`. (#6373)
* [changed] **Breaking Change**: Migrated `FirebaseVertexAIException` from a sealed class to an abstract class, and marked constructors as internal. (#6368)
* [feature] Added support for `title` and `publicationDate` in citations. (#6309)
* [feature] Added support for `frequencyPenalty`, `presencePenalty`, and `HarmBlockMethod`. (#6309)
* [changed] **Breaking Change**: Introduced `Citations` class. Now `CitationMetadata` wraps that type. (#6276)
* [changed] **Breaking Change**: Reworked `Schema` declaration mechanism. (#6258)
* [changed] **Breaking Change**: Reworked function calling mechanism to use the new `Schema` format. Function calls no longer use native types, nor include references to the actual executable code. (#6258)
* [changed] **Breaking Change**: Made `totalBillableCharacters` field in `CountTokens` nullable and optional. (#6294)
* [changed] **Breaking Change**: Removed `UNKNOWN` option for the `HarmBlockThreshold` enum. (#6294)
* [changed] **Breaking Change**: Removed `UNSPECIFIED` option for the `HarmBlockThreshold`, `HarmProbability`, `HarmSeverity`, and `BlockReason` enums. (#6294)
* [changed] **Breaking Change**: Renamed `BlockThreshold` as `HarmBlockThreshold`. (#6262)
* [changed] **Breaking Change**: Renamed all types and methods starting with `blob` to start with `inlineData`. (#6309)
* [changed] **Breaking Change**: Changed the order of arguments in `InlineDataPart` to match `ImagePart`. (#6340)
* [changed] **Breaking Change**: Changed `RequestOption` to accept only `long` timeout values. (#6289)
* [changed] **Breaking Change**: Moved `requestOptions` to the last positional argument in the `generativeModel` argument list. (#6292)
* [changed] **Breaking Change**: Replaced sealed classes with abstract classes for `StringFormat`. (#6334)
* [changed] **Breaking Change**: Refactored enum classes to be normal classes. (#6340)
* [changed] **Breaking Change**: Marked `GenerativeModel` properties as private. (#6309)
* [changed] **Breaking Change**: Changed `method` parameter type to be nullable in `SafetySettings`. (#6379)

# 16.0.0-beta05
* [changed] Merged core networking code into VertexAI from a separate library
* [feature] added support for `responseSchema` in `GenerationConfig`.

# 16.0.0-beta03
* [changed] Breaking Change: changed `Schema.int` to return 32 bit integers instead of 64 bit (long).
* [changed] Added `Schema.long` to return 64-bit integer numbers.
* [changed] Added `Schema.double` to handle floating point numbers.
* [changed] Marked `Schema.num` as deprecated, prefer using `Schema.double`.
* [fixed] Fixed an issue with decoding JSON literals (#6028).

# 16.0.0-beta01
* [feature] Added support for `responseMimeType` in `GenerationConfig`.
* [changed] Renamed `GoogleGenerativeAIException` to `FirebaseVertexAIException`.
* [changed] Updated the KDocs for various classes and functions.

Loading
Loading