Interface BatchEmbeddingModel

All Known Implementing Classes:
GoogleAiGeminiBatchEmbeddingModel
@Experimental public interface BatchEmbeddingModel
Used for processing multiple embedding requests asynchronously in a batch.

Batch processing typically offers significant cost reductions compared to real-time requests and is ideal for large-scale, non-urgent tasks.

Each successful result is a Response wrapping the computed Embedding. The Response carries the per-embedding TokenUsage when the provider reports it, and null otherwise.

See Also:

  • Method Details

    • submit

      BatchResponse<Response<Embedding>> submit(BatchRequest<TextSegment> request)
      Creates a batch of text segments and submits them for asynchronous embedding processing.

      The returned BatchResponse represents the status of the batch operation.

      Parameters:
      request - the list of text segments to embed in the batch
      Returns:
      a BatchResponse representing the initial state of the batch operation

    • retrieve

      BatchResponse<Response<Embedding>> retrieve(String batchId)
      Retrieves the current state and results of an embedding batch operation.

      The response indicates whether the batch is still processing, completed successfully, or failed. Use this to retrieve the computed embeddings once the state is success.

      Parameters:
      batchId - the batch identifier obtained from submit(BatchRequest)
      Returns:
      a BatchResponse representing the current state of the embedding batch operation

    • cancel

      void cancel(String batchId)
      Cancels an embedding batch operation that is currently pending or running.
      Parameters:
      batchId - the batch identifier to cancel

    • list

      BatchPage<Response<Embedding>> list(@Nullable BatchPagination pagination)
      Lists embedding batch jobs with optional pagination.
      Parameters:
      pagination - the maximum number of batch jobs to return and token for retrieving a specific page; if null, uses server default
      Returns:
      a BatchPage containing embedding batch responses and pagination information