AlloyDB AI adds pre-registered support for the following Gemini models in version
Share
Services
## Feature
Feature
AlloyDB AI adds pre-registered support for the following Gemini models in version 1.6.5 and later of the `google_ml_integration` extension:
* `gemini-3.1-flash-lite`
* `gemini-3.5-flash`
* `gemini-3.5-flash-lite`
* `gemini-3.6-flash`
* `gemini-3.7-flash`
* `gemini-3.8-flash`
* `gemini-4-argon`
Version 1.6.5 of the `google_ml_integration` extension sets `gemini-3.5-flash-lite` as the default Gemini model and removes the `gemini-2.0-flash`, `gemini-2.0-flash-lite`, `gemini-2.5-flash`, `gemini-2.5-flash-lite`, and `gemini-2.5-pro` models from the list of pre-registered models.
Additionally, AlloyDB AI adds pre-registered support for the `semantic-ranker-512` reranking model.
For more information, see[Pre-registered Agent Platform models](https://cloud.google.com/alloydb/docs/reference/supported-ai-models#pre-registered-models).
What else is happening at Google Cloud Platform?
Read update
Services
Share
Read update
Services
Share
Read update
Services
Share
Managed OpenTelemetry for GKE is generally available (GA) on clusters
about 13 hours ago
Services
Share
Dataflow has added the following features to tune horizontal
about 19 hours ago
Services
Share