Packages

@ai-sdk/google-vertex

5.0.144 · 20 Nov 2026Apache-2.06.4M / weekRepositoryHomepagenpmAtom feed

Your version

682 stable releases, newest first, by release line. Notes are quoted as the maintainers published them.

  1. 5.0.144

    20 Nov 2026
    • google, google-vertex: Finish live transcription streams without the 3-second grace period

      gemini-3.5-transcribe-live ends each segment with generationComplete and never sends turnComplete, so live transcription streams always waited out the 3-second finish grace period and merged all segments into one transcript-final without separating whitespace. Segments now complete on generationComplete, and streams finish as soon as server voice activity detection shows that all sent audio has been transcribed.

      Adds providerOptions.google.streaming.automaticActivityDetection.disabled (googleVertex on Vertex) for push-to-talk input: the audio stream is transcribed as a single turn without splitting at pauses, and the transcript is finalized as soon as the audio input closes.

  2. 5.0.143

    20 Nov 2026
    • google: Add the gemini-4-neon model ID with xhigh thinking support

  3. 5.0.142

    17 Nov 2026
    • provider/google: Redact MCP server credentials in telemetry

      Expose googleVertex.tools.mcpServer for Google Vertex Interactions models. Its
      MCP server header values are redacted from telemetry, matching
      @ai-sdk/google.

  4. 5.0.141

    13 Nov 2026
    • Derive known model ID suggestions from each provider's model catalog declarations so catalog updates automatically update autocomplete, while preserving existing suggestions, endpoint-specific model lists, and support for custom model IDs.

    • Add the missing gemini-4-hydrogen model ID to the Vertex catalog so model suggestions and generated capability documentation include it.

  5. 5.0.139

    12 Nov 2026
    • Forward the provider's configured fetch transport when downloading file inputs. Add a fetch option to createDownload so core downloads can also use guarded-fetch for opt-in DNS validation on Node.js. The function returned by createDownload can now also be passed as experimental_download to generateText and streamText, and experimental_download now also handles generated file URLs returned by the model, so one hook covers all core downloads. Document that the default URL checks do not resolve DNS and recommend the guarded transport or network egress controls for untrusted URLs.

      Keep Azure image and Vertex tool-result downloads separate from provider authentication wrappers so file hosts never receive Azure bearer tokens or Vertex Express Mode API keys.

      Pass the call's abort signal to custom download requests, including prompt assets and generated files, so experimental_download: createDownload({ fetch }) preserves cancellation and timeouts.

  6. 5.0.138

    11 Nov 2026

    Only updates a dependency:@ai-sdk/google 4.0.118

  7. 5.0.137

    10 Nov 2026
    • Add support for the gemini-4-carbon model with the same capabilities as gemini-4-argon.

  8. 5.0.136

    10 Nov 2026
    • acdccac: Route off-origin MaaS image downloads through the validated download transport so the built-in authentication wrappers in both entry points preserve Node.js DNS validation and manual redirect handling. Explicit custom fetch implementations and the browser redirect fallback remain supported.

  9. 5.0.134

    7 Nov 2026
    • Pass providerOptions.google.labels through to Google and Vertex AI Interactions requests, including streaming, so callers can attach request metadata such as safety_identifier.

    • Only send google cloud credentials from the maas provider to its own api, not to image urls from a response or their redirects

    • Omit Google provider-defined tools, including File Search and Vertex RAG Store, when toolChoice is none so disabled tools cannot retrieve data or execute server-side. Preserve function declarations needed by conversation history while disabling function calling. This also applies when structured output uses a synthetic JSON response tool.

  10. 5.0.133

    6 Nov 2026
    • Refactor Google Vertex model ID logic into a model catalog that declares capabilities and provider options, and generate provider capability documentation from it. Export Google catalog metadata internally for Vertex to reuse. Request behavior is unchanged.

  11. 5.0.132

    5 Nov 2026
    • Preserve the order of text and complete function calls in streamed Google responses, including standalone thought signatures on empty text or reasoning parts.

      Keep provider metadata on empty text parts when converting results to response messages so it can be passed back in subsequent requests.

      Omit foreign-provider empty signature text when replaying history through Anthropic or Bedrock, preserving Bedrock’s existing native reasoning-history handling.

    • Pass through seed: 0 instead of dropping it in image and video models. KlingAI now reports seed: 0 as an unsupported setting like any other seed.

    • Expose optional transcription word timings separately from segments and accept shared timestampGranularities. Explicit provider timing settings take precedence. OpenAI now sends timestamp_granularities only with verbose_json and warns when a requested timing is unavailable or excluded.

  12. 5.0.129 – 5.0.1313 releases that only update dependencies@ai-sdk/provider-utils 5.0.74@ai-sdk/google 4.0.111@ai-sdk/anthropic 4.0.96and 1 moreShow each
  13. 5.0.128

    30 Oct 2026
    • Support promises and synchronous or asynchronous callbacks in provider API key and token settings. Resolve credentials per authenticated request while preserving existing authentication behavior and synchronous credential serialization.

      The experimental OpenAI Live getServerWebSocketConfig() method can now resolve asynchronously. Await its result before using the connection settings.

  14. 5.0.126 – 5.0.1272 releases that only update dependencies@ai-sdk/google 4.0.108@ai-sdk/provider-utils 5.0.71@ai-sdk/anthropic 4.0.92and 1 moreShow each
  15. 5.0.125

    27 Oct 2026
  16. 5.0.120 – 5.0.1245 releases that only update dependencies@ai-sdk/anthropic 4.0.90@ai-sdk/provider-utils 5.0.70@ai-sdk/google 4.0.107and 2 moreShow each
  17. 5.0.119

    21 Oct 2026
    • Defer construction of provider validation schemas until first use to reduce package import time.

  18. 5.0.116 – 5.0.1183 releases that only update dependencies@ai-sdk/anthropic 4.0.84@ai-sdk/provider 4.0.31@ai-sdk/provider-utils 5.0.64and 2 moreShow each
  19. 5.0.115

    16 Oct 2026
    • google: Surface in-stream errors

  20. 5.0.113 – 5.0.1142 releases that only update dependencies@ai-sdk/google 4.0.98@ai-sdk/provider 4.0.28@ai-sdk/openai-compatible 3.0.70and 2 moreShow each
  21. 5.0.112

    15 Oct 2026
    • vertex-anthropic: Forward 'providerOptions.vertex.labels' as a header

  22. 5.0.111

    13 Oct 2026
    • google: Add gemini-4-argon model id and long decode support for generateContent: requests that reach the per-request cap (CONTINUATION) are continued automatically with usage summed across requests and the first request's prompt token count exposed as providerMetadata.google.initialPromptTokenCount, and the continuationToken provider option / provider metadata resume responses that used up their maxOutputTokens budget. generateText with the Interactions API now polls background: true model calls until they finish.

  23. 5.0.109

    8 Oct 2026
    • Map the top level embedding dimensions to provider settings

  24. 5.0.106 – 5.0.1083 releases that only update dependencies@ai-sdk/anthropic 4.0.77@ai-sdk/provider 4.0.25@ai-sdk/provider-utils 5.0.57and 2 moreShow each
  25. 5.0.105

    7 Oct 2026
    • anthropic: Add Claude Haiku 5.5 support

  26. 5.0.103

    6 Oct 2026
    • google: Add the 'gemini-nano-banana-2.1' model ID

  27. 5.0.101 – 5.0.1022 releases that only update dependencies@ai-sdk/provider 4.0.22@ai-sdk/provider-utils 5.0.54@ai-sdk/anthropic 4.0.72and 2 moreShow each
  28. 5.0.100

    30 Sep 2026
    • ai: Add telemetry to speech generation and transcription, including provider usage propagation and experimental streaming lifecycle callbacks

    • Validate hostname parts using isValidHostnamePart

  29. 5.0.98

    28 Sep 2026
    • Use standards-compliant User-Agent header

  30. 5.0.97

    28 Sep 2026
    • anthropic: Add Claude Sonnet 5.5 support

  31. 5.0.96

    28 Sep 2026
    • provider: Advertise image model file and mask input support

      Use confirmed model IDs for capability declarations so unrecognized model names remain unknown. Include Together AI FLUX.2 Pro and Flex single-image editing, and allow asynchronous capability lookups and middleware overrides to resolve to unknown.

      Advertise QuiverAI Arrow 2 and Arrow 2 Telos file-input support, and mark Together AI Gemini image inputs unsupported by the current single-image request mapping.

  32. 5.0.95

    25 Sep 2026
    • google: Forward supported Vertex tool result URLs as function response file data

  33. 5.0.93

    24 Sep 2026

A whole project to check? Paste its package.json in the dependency check: every deprecated, outdated or vulnerable dependency, with what to upgrade to. Free, no account.