[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-04-03。"],[[["AlloyDB AI integrates machine learning capabilities into AlloyDB for PostgreSQL and AlloyDB Omni, allowing users to apply ML models to their data."],["The `vector` extension, a customized version of `pgvector`, is available for storing embeddings in vector columns and creating `IVF`, `IVFFlat`, or `HSNW` indexes."],["AlloyDB AI introduces the `alloydb_scann` extension, which implements a highly efficient nearest-neighbor index using the ScaNN algorithm for PostgreSQL 15 compatible databases."],["Users can utilize the `google_ml_integration` extension to generate embeddings and invoke predictions through SQL transactions, using functions like `embedding()` and `google_ml.embedding()`."],["AlloyDB Omni can be configured to integrate with Vertex AI, enabling applications to invoke predictions from any model in the Vertex AI Model Garden and generate embeddings using the `text-embedding-005` English models."]]],[]]