Pinecone TypeScript SDK - v9.0.0
    Preparing search index...

    Interface SemanticTextField

    A semantic text field, as returned when describing an index created with an integrated embedding model. Backed by an integrated embedding model that embeds text at write and query time, enabling semantic similarity search without separate embedding calls. Cannot be declared when creating an index; use the create-for-model operation.

    SemanticTextField

    interface SemanticTextField {
        description?: string | null;
        dimension?: number | null;
        metric?: string;
        model: string;
        readParameters?: object | null;
        type: "semantic_text";
        writeParameters?: object | null;
    }
    Index
    description?: string | null

    Optional description for this field, at most 256 bytes. null in responses when none was set.

    SemanticTextField

    dimension?: number | null

    The dimension of the vectors the model produces. null for models that produce sparse vectors.

    SemanticTextField

    metric?: string

    The distance metric used for similarity search. Defaults to the model's preferred metric if not specified. Possible values: cosine, dotproduct, or euclidean.

    SemanticTextField

    model: string

    The name of the integrated embedding model to use for this field.

    SemanticTextField

    readParameters?: object | null

    Model-specific parameters applied at query time, such as input_type.

    SemanticTextField

    type: "semantic_text"

    Identifies this as a semantic text field. Must be semantic_text.

    SemanticTextField

    writeParameters?: object | null

    Model-specific parameters applied at write time, such as input_type.

    SemanticTextField