Starting April 29, 2025, Gemini 1.5 Pro and Gemini 1.5 Flash models are not available in projects that have no prior usage of these models, including new projects. For details, see Model versions and lifecycle.
Perform an online explanation.
If
deployed_model_id
is specified, the corresponding DeployModel must have
explanation_spec
populated. If
deployed_model_id
is not specified, all DeployedModels must have
explanation_spec
populated.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-16 UTC."],[],[],null,["# Interface PredictionServiceGrpc.AsyncService (1.32.0)\n\n public static interface PredictionServiceGrpc.AsyncService\n\nA service for online predictions and explanations.\n\nMethods\n-------\n\n### directPredict(DirectPredictRequest request, StreamObserver\\\u003cDirectPredictResponse\\\u003e responseObserver)\n\n public default void directPredict(DirectPredictRequest request, StreamObserver\u003cDirectPredictResponse\u003e responseObserver)\n\nPerform an unary online prediction request to a gRPC model server for\nVertex first-party products and frameworks.\n\n### directRawPredict(DirectRawPredictRequest request, StreamObserver\\\u003cDirectRawPredictResponse\\\u003e responseObserver)\n\n public default void directRawPredict(DirectRawPredictRequest request, StreamObserver\u003cDirectRawPredictResponse\u003e responseObserver)\n\nPerform an unary online prediction request to a gRPC model server for\ncustom containers.\n\n### explain(ExplainRequest request, StreamObserver\\\u003cExplainResponse\\\u003e responseObserver)\n\n public default void explain(ExplainRequest request, StreamObserver\u003cExplainResponse\u003e responseObserver)\n\nPerform an online explanation.\nIf\ndeployed_model_id\nis specified, the corresponding DeployModel must have\nexplanation_spec\npopulated. If\ndeployed_model_id\nis not specified, all DeployedModels must have\nexplanation_spec\npopulated.\n\n### generateContent(GenerateContentRequest request, StreamObserver\\\u003cGenerateContentResponse\\\u003e responseObserver)\n\n public default void generateContent(GenerateContentRequest request, StreamObserver\u003cGenerateContentResponse\u003e responseObserver)\n\nGenerate content with multimodal inputs.\n\n### predict(PredictRequest request, StreamObserver\\\u003cPredictResponse\\\u003e responseObserver)\n\n public default void predict(PredictRequest request, StreamObserver\u003cPredictResponse\u003e responseObserver)\n\nPerform an online prediction.\n\n### rawPredict(RawPredictRequest request, StreamObserver\\\u003cHttpBody\\\u003e responseObserver)\n\n public default void rawPredict(RawPredictRequest request, StreamObserver\u003cHttpBody\u003e responseObserver)\n\nPerform an online prediction with an arbitrary HTTP payload.\nThe response includes the following HTTP headers:\n\n- `X-Vertex-AI-Endpoint-Id`: ID of the Endpoint that served this prediction.\n- `X-Vertex-AI-Deployed-Model-Id`: ID of the Endpoint's DeployedModel that served this prediction.\n\n### serverStreamingPredict(StreamingPredictRequest request, StreamObserver\\\u003cStreamingPredictResponse\\\u003e responseObserver)\n\n public default void serverStreamingPredict(StreamingPredictRequest request, StreamObserver\u003cStreamingPredictResponse\u003e responseObserver)\n\nPerform a server-side streaming online prediction request for Vertex\nLLM streaming.\n\n### streamDirectPredict(StreamObserver\\\u003cStreamDirectPredictResponse\\\u003e responseObserver)\n\n public default StreamObserver\u003cStreamDirectPredictRequest\u003e streamDirectPredict(StreamObserver\u003cStreamDirectPredictResponse\u003e responseObserver)\n\nPerform a streaming online prediction request to a gRPC model server for\nVertex first-party products and frameworks.\n\n### streamDirectRawPredict(StreamObserver\\\u003cStreamDirectRawPredictResponse\\\u003e responseObserver)\n\n public default StreamObserver\u003cStreamDirectRawPredictRequest\u003e streamDirectRawPredict(StreamObserver\u003cStreamDirectRawPredictResponse\u003e responseObserver)\n\nPerform a streaming online prediction request to a gRPC model server for\ncustom containers.\n\n### streamGenerateContent(GenerateContentRequest request, StreamObserver\\\u003cGenerateContentResponse\\\u003e responseObserver)\n\n public default void streamGenerateContent(GenerateContentRequest request, StreamObserver\u003cGenerateContentResponse\u003e responseObserver)\n\nGenerate content with multimodal inputs with streaming support.\n\n### streamRawPredict(StreamRawPredictRequest request, StreamObserver\\\u003cHttpBody\\\u003e responseObserver)\n\n public default void streamRawPredict(StreamRawPredictRequest request, StreamObserver\u003cHttpBody\u003e responseObserver)\n\nPerform a streaming online prediction with an arbitrary HTTP payload.\n\n### streamingPredict(StreamObserver\\\u003cStreamingPredictResponse\\\u003e responseObserver)\n\n public default StreamObserver\u003cStreamingPredictRequest\u003e streamingPredict(StreamObserver\u003cStreamingPredictResponse\u003e responseObserver)\n\nPerform a streaming online prediction request for Vertex first-party\nproducts and frameworks.\n\n### streamingRawPredict(StreamObserver\\\u003cStreamingRawPredictResponse\\\u003e responseObserver)\n\n public default StreamObserver\u003cStreamingRawPredictRequest\u003e streamingRawPredict(StreamObserver\u003cStreamingRawPredictResponse\u003e responseObserver)\n\nPerform a streaming online prediction request through gRPC."]]
