$vectorSearch (aggregation stage)（聚合阶段）

Definition定义

$vectorSearch

$vectorSearch performs a semantic search on data in your Atlas cluster.对Atlas集群中的数据执行语义搜索。

If you store vector embeddings on your Atlas cluster, you can seamlessly index the vector data along with your other collection data. 如果将向量嵌入存储在Atlas集群上，则可以将向量数据与其他集合数据无缝地索引。You can use the $vectorSearch stage to pre-filter your data and perform semantic search against the indexed fields.您可以使用$vectorSearch阶段预筛选数据，并对索引字段执行语义搜索。

Your vector data must be less than or equal to 4096 dimensions in width.矢量数据的宽度必须小于或等于4096维。

Behavior行为

Starting in MongoDB 8.0, you can use a $vectorSearch stage in a $unionWith stage.从MongoDB 8.0开始，您可以在$unionWith阶段中使用$vectorSearch阶段。

Limitations局限性

You cannot use a $vectorSearch stage in a $facet stage or a $lookup stage.您不能在$facet阶段或$lookup阶段中使用$vectorSearch阶段。

Learn More了解更多

• To learn more about creating Vector Search indexes, see Index Vector Embeddings.要了解有关创建向量搜索索引的更多信息，请参阅索引矢量嵌入。
• To learn more about $vectorSearch pipeline stage syntax and usage, see Vector Search Queries.要了解有关$vectorSearch管道阶段语法和用法的更多信息，请参阅矢量搜索查询。