[renovate]

This PR contains the following updates:

Package	Change	Age	Confidence
@google/genai	2.19.0 → 3.2.0

---

Release Notes

googleapis/js-genai (@​google/genai)

v3.2.0

Compare Source

Features

• add the Decisions API (client.decisions.create) to the GenAI SDKs (3712be2)
• environments: introduce self-hosted BYOS mode, system_instruction, custom_config, and environment_key in SDKs and CLI (ac855bd)

v3.1.0

Compare Source

Features

• add gemini-omni-1.2-flash and make it the gemini-api video default (cd31e9f)
• Add proactive_video to ProactivityConfig for the Live API. (2164fe7)
• add xhigh thinking level to Interactions ThinkingLevel (eff6729)
• add usage_metadata (including traffic_type) and truncated to EmbedContentResponse for Gemini Enterprise API (81d42ca)
• interactions: add gemini-4-carbon and gemini-omni-1.2-flash to Model enum (d406013)

v3.0.0

Compare Source

⚠ BREAKING CHANGES

• Default Gemini Developer API version is now v1 (was v1beta) (3f477bf)

  • Clients targeting the Gemini Developer API now default to API version v1 instead of v1beta.
  • No effect on Gemini Enterprise (Vertex AI) defaults: Clients with vertexai: true are unaffected by this default change and continue to default to v1beta1.
  • Some models are not available on v1, notably: the gemini-flash-latest, gemini-flash-lite-latest and gemini-pro-latest aliases, and some -preview models. They exist only on v1beta. To keep using them, use a specific stable model version, or switch back to v1beta as shown below.
  • How to configure or restore the previous default via environment variable (Node.js): Set GOOGLE_GENAI_API_VERSION (works for both Gemini Developer API and Enterprise / Vertex AI clients when no version is set in code):

    # Restores the beta default ('v1beta' on Gemini Developer API, 'v1beta1' on Enterprise / Vertex AI)
    export GOOGLE_GENAI_API_VERSION=beta
    
    # Or set an explicit API version string on either platform (e.g. 'v1', 'v1beta', 'v1beta1')
    export GOOGLE_GENAI_API_VERSION=v1beta

  • How to set the API version in code: Pass apiVersion (or httpOptions.apiVersion) to GoogleGenAI (takes precedence over GOOGLE_GENAI_API_VERSION):

    import { GoogleGenAI } from '@google/genai';
    
    const ai = new GoogleGenAI({
      apiVersion: 'v1beta',
    });

• Minimum supported Node.js version is now 22 (was 20) (7b9fbf0)

  • Following Node.js 20 reaching end-of-life (2026-04-30), package.json engines.node now requires >=22.0.0.
  • There are no runtime API or bundle changes resulting from this bump, and both CommonJS and ESM bundles continue to be published.
  • Users who must remain on Node.js 20 should pin to @google/genai@^2 (or ignore engine checks if supported by their environment).

• interactions: CodeExecutionResultStep.result is now typed as Array<Content> | string (CodeExecutionResultStepResult) instead of string (b36ce99). The JSON and runtime value for text-only results are still a plain string, but TypeScript code reading step.result as a string may need to narrow the union (typeof step.result === 'string').

Features

• Add source_text to AudioTranscriptionConfig and SmartTranscriptionMode. (ddb5158)
• Add MALFORMED_RESPONSE to FinishReason (032d50c), refs #​2937
• cli: use gemini-3.5-transcribe as default model for transcribe command (df29944)
• default Gemini API version to v1 instead of v1beta across SDKs (3f477bf)
• Disable GenerateVideos for Gemini API, update tests/READMEs (3f65d02)
• drop support for Node 20; Node.js >=22 is now required (7b9fbf0)
• interactions: add Office and OpenDocument MIME types (b24cfc2)
• interactions: support multimodal content in CodeExecutionResultStep.result (b36ce99)
• support GOOGLE_GENAI_API_VERSION environment variable across SDKs (3f477bf)
• update discovery doc (8be9344)

Bug Fixes

• Raise protobufjs and ws minimums to patched versions (71827d0), refs #​1611
• Reject Live.connect() when the WebSocket errors or closes before setup completes (04eaa81), refs #​1257
• respect configured and default API version for Gemini API file uploads across SDKs (bdb4e42)
• stop emitting array indices in auth token fieldMask (80e7b3d), refs #​1492

v2.32.0

Compare Source

Features

• Add support for continuation_token and requires_continuation status (c7b2fd2)
• add video upscale task parameter (54b9b19)
• interactions: add automatic_continuation to interactions.create in the TypeScript SDK (99f1892)
• interactions: declare continuation_ids on Interaction (9cf3280)
• support video understanding in Gemini Developer API (7c82322)

Bug Fixes

• interactions: align SDK User-Agent and x-goog-api-client headers (dda511e)

v2.31.0

Compare Source

Features

• Add store to GenerateContentConfig (Gemini Developer API only) and labels to LiveConnectConfig. (0b52aa5)
• add base_voice to PromptedVoice in GAOS SDK (d62820b)
• Add generation_output_config to GenerateContentConfig (e82a75a)
• Add Live API in-session context updates (send_context_update) and types across GenAI SDKs. (177ee68)
• include AudioContent, DocumentContent, and VideoContent in FunctionResultSubContent (aeaed4e)
• sdks: enable fixes.openSingleValueEnumsOct2026 across all SDKs (0059474)
• update discovery doc (40d6e98)

Bug Fixes

• genai: raise mid-stream SSE errors and dropped connections consistently across SDKs; add continuation conformance tests (38f6ebc)
• interactions: sync InteractionSseEventInteractionStatus enum with InteractionStatus (be6aa9f)
• triggers: fix agent triggers OpenAPI overlays, numeric fields, and SDK patches (0e9031d)

v2.30.0

Compare Source

Features

• interactions: add gemini-4-argon to Model enum (8a3c307)

Bug Fixes

• clear continuation state and preserve prompt feedback when continuation hop returns no candidates (e63d1f1)
• fix the grounding metadata merging for automatic continuation in generate content and generate content stream (6bfba87)
• make continuation streams retry dropped hops, stop on prompt_feedback, and keep real errors in generate_content (6bdc0d9)
• type GeminiNextGen signatures with public, importable names (ad3fb89)

v2.29.1

Compare Source

Bug Fixes

• concatenate output_text across multi-step model responses in a turn (567cc8a)

v2.29.0

Compare Source

Features

• accumulate streaming continuation usage metadata and expose initialPromptTokenCount (dbc961f)
• Add continuation_token to the Interactions interaction.status_update event. (a93776c)
• Add a CONTINUATION_REQUIRED Interaction status (754f71f)
• add billable token counts fields (839ffc2)
• Add startOffset/endOffset to Transcription and interimTranscriptTimestampEnabled to RealtimeInputConfig (bdbf156)
• Add usage in InteractionStatusUpdate events (8892aea)
• automatically consume the continuation token (98637d1)
• cli: update Speakeasy generator to v2.948.0, bump CLI to v0.4.2, and apply CLI overlay fixes (22d53ec)

Bug Fixes

• clear intermediate CONTINUATION finishReason in streaming continuation (22706af)
• do not record incomplete continuation stream turns in curated chat history (81c3c49)
• gemini-api-cli: remove unsupported 2.5 and 1.6 models from Interactions model options and CLI examples (1002720)
• js: do not set initial prompt token count for requests that do not use automatic consumption of continuation token (0885fb9)
• js: fix streaming continuation edge cases (stale token, AbortError, initialPromptTokenCount) (f3d10a6)
• Raise default HTTP timeouts so long-running requests aren't killed client-side (10e850c)
• resume streaming generation from the latest checkpoint continuation token on mid-stream errors or cutoffs (625ac2c)
• Update EnvironmentConfig.env schema to map keys to EnvVar or string. (8478e12)

v2.28.0

Compare Source

Features

• Support "allowlist": "disabled" in environment network config. (18bc392)
• update discovery doc (4e996fe)

v2.27.0

Compare Source

Features

• add lyria-3.5, gemini-omni-1.1-flash, and gemini-omni-flash-preview to Interactions Model enum (4ce63d1)
• support continuation_token in Interactions (a888ad7)

v2.26.0

Compare Source

Features

• support continuation_token in GenerateContent (ac628c5)

v2.25.0

Compare Source

Features

• add gcs_metrics_uri for exporting tuning job metrics. (bc085fb)
• Add Gemini 3.8 Flash TTS and Gemini 3.8 Flash Lite TTS models to SDKs (b12dad0)
• genai-sdk: Add environments dedicated module and from_environment support (4d7e80b)
• include labels for LiveClientSetup (d4bcdad)
• support labels for Gemini API GenerateContent (71c3997)

Bug Fixes

• use CreateAgentInteraction for Trigger.interaction and TriggerCreateParams.interaction (c4341e1)

v2.24.0

Compare Source

Features

• add sample_audio to Voice in GAOS SDK (d001d18)
• add Voices API resource to GAOS SDK (f97113f)
• expose SpeechMetadata, VoiceConfig.voice, and SpeechAnnotation in public GenAI SDKs (19b7858)
• support field.value_field syntax for map add_union_value annotations. (51d701a)
• update discovery doc (42f9716)
• wire voice into sdk (08b568e)

Bug Fixes

• sanitize path parameter descriptions in GAOS OpenAPI generation (1567962)

v2.23.0

Compare Source

Features

• add credential APIs resource to gaos. (0e6410b)
• Add RetrievalCallStep and RetrievalResultStep to interactions schema and SDKs. (d03d569)
• implement environments files upload and download across Python and TypeScript SDKs (3a4c512)
• support custom fetch in HttpOptions (01897b3)
• wire credential APIs resource to sdk. (280da75)

v2.22.0

Compare Source

Features

• Add from_environment support for environment copying (01ed17e)

Bug Fixes

• Add @​ts-ignore before optional MCP imports in declaration files (7b5ef39)
• Remove unused field from FunctionResultDelta (aeb7453)
• Repair node fallback manifest and complete web entries (0f59bf8)

v2.21.0

Compare Source

Features

• Add Gemini 3.8 Flash model to SDKs and update Flash model descriptions (4a969fe)

v2.20.0

Compare Source

Features

• Add support for the audio/webm MIME type. (b0e3c5f)
• Add translation_config SDK support for GEAP. (3e1d923)
• add Video Understanding support to the Interactions API (6ddbeba)

Bug Fixes

• Don't set redundant Content-Length on file upload requests (2f5cc7a)
• Don't set redundant Content-Length on file upload requests (2f5cc7a), refs #​1718
• expose ProcessingCallStep and ProcessingResultStep in Interactions SDK (b6f8db5)
• Quote setupComplete key in LiveClientMessage (4467faf)

---

Configuration

📅 Schedule: (in timezone Asia/Taipei)

• Branch creation
  • Between 12:00 AM and 05:59 AM, only on Monday (* 0-5 * * 1)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
chia1104	Error		Nov 15, 2026 6:31pm UTC

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: defaults
• Review profile: CHILL
• Plan: Advanced
• Run ID: fd614c4a-187b-4947-b203-70bdb1c1f2af

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• Trigger review

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

Share

• X
• Mastodon
• Reddit
• LinkedIn