The dimension of the embedding vectors.
The distance metric used for similarity search.
Possible values: cosine, dotproduct, or euclidean.
The hosted dense embedding model that embeds the field's text and the query text of embed scoring methods.
OptionalreadThe model-specific parameters applied when embedding the query text of embed scoring methods, as resolved at creation: the parameters set in the create request merged with the model's defaults.
OptionalwriteThe model-specific parameters applied when embedding document text, as resolved at creation: the parameters set in the create request merged with the model's defaults.
The dense embedding configuration of a
stringfield, as resolved when the index was created: the values set in the create request, with every omitted one filled from the model's defaults. Present when the field was created withembed.Export
ResponseEmbedConfig