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

    Interface CreateIndexForModelEmbed

    The integrated embedding configuration for a new index.

    interface CreateIndexForModelEmbed {
        dimension?: number;
        fieldMap: Record<string, string>;
        metric?: IndexMetric;
        model: string;
        readParameters?: object;
        writeParameters?: object;
    }

    Hierarchy

    • Omit<CreateIndexForModelRequestEmbed, "metric" | "fieldMap">
      • CreateIndexForModelEmbed
    Index
    dimension?: number

    The dimension of embedding vectors produced for the index.

    CreateIndexForModelRequestEmbed

    fieldMap: Record<string, string>

    Maps the model's input to a field of your documents. Pass { text: 'chunk_text' } to embed the chunk_text field of every document you upsert.

    metric?: IndexMetric

    The distance metric to use for similarity search. Defaults to the model's preferred metric.

    model: string

    The embedding model to use. Call Inference.listModels to find a model suited to your document text and search queries.

    readParameters?: object

    The read parameters for the embedding model.

    CreateIndexForModelRequestEmbed

    writeParameters?: object

    The write parameters for the embedding model.

    CreateIndexForModelRequestEmbed