google-cloud-gke_recommender-v1 0.a → 0.1.0

data/proto_docs/google/cloud/gkerecommender/v1/gkerecommender.rb

@@ -0,0 +1,558 @@
+# frozen_string_literal: true
+
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Auto-generated by gapic-generator-ruby. DO NOT EDIT!
+
+
+module Google
+  module Cloud
+    module GkeRecommender
+      module V1
+        # Request message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_models GkeInferenceQuickstart.FetchModels}.
+        # @!attribute [rw] page_size
+        #   @return [::Integer]
+        #     Optional. The target number of results to return in a single response.
+        #     If not specified, a default value will be chosen by the service.
+        #     Note that the response may include a partial list and a caller should
+        #     only rely on the response's
+        #     {::Google::Cloud::GkeRecommender::V1::FetchModelsResponse#next_page_token next_page_token}
+        #     to determine if there are more instances left to be queried.
+        # @!attribute [rw] page_token
+        #   @return [::String]
+        #     Optional. The value of
+        #     {::Google::Cloud::GkeRecommender::V1::FetchModelsResponse#next_page_token next_page_token}
+        #     received from a previous `FetchModelsRequest` call.
+        #     Provide this to retrieve the subsequent page in a multi-page list of
+        #     results. When paginating, all other parameters provided to
+        #     `FetchModelsRequest` must match the call that provided the page token.
+        class FetchModelsRequest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Response message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_models GkeInferenceQuickstart.FetchModels}.
+        # @!attribute [r] models
+        #   @return [::Array<::String>]
+        #     Output only. List of available models. Open-source models follow the
+        #     Huggingface Hub `owner/model_name` format.
+        # @!attribute [r] next_page_token
+        #   @return [::String]
+        #     Output only. A token which may be sent as
+        #     [page_token][FetchModelsResponse.page_token] in a subsequent
+        #     `FetchModelsResponse` call to retrieve the next page of results.
+        #     If this field is omitted or empty, then there are no more results to
+        #     return.
+        class FetchModelsResponse
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Request message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_servers GkeInferenceQuickstart.FetchModelServers}.
+        # @!attribute [rw] model
+        #   @return [::String]
+        #     Required. The model for which to list model servers. Open-source models
+        #     follow the Huggingface Hub `owner/model_name` format. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_models GkeInferenceQuickstart.FetchModels}
+        #     to find available models.
+        # @!attribute [rw] page_size
+        #   @return [::Integer]
+        #     Optional. The target number of results to return in a single response.
+        #     If not specified, a default value will be chosen by the service.
+        #     Note that the response may include a partial list and a caller should
+        #     only rely on the response's
+        #     {::Google::Cloud::GkeRecommender::V1::FetchModelServersResponse#next_page_token next_page_token}
+        #     to determine if there are more instances left to be queried.
+        # @!attribute [rw] page_token
+        #   @return [::String]
+        #     Optional. The value of
+        #     {::Google::Cloud::GkeRecommender::V1::FetchModelServersResponse#next_page_token next_page_token}
+        #     received from a previous `FetchModelServersRequest` call.
+        #     Provide this to retrieve the subsequent page in a multi-page list of
+        #     results. When paginating, all other parameters provided to
+        #     `FetchModelServersRequest` must match the call that provided the page
+        #     token.
+        class FetchModelServersRequest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Response message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_servers GkeInferenceQuickstart.FetchModelServers}.
+        # @!attribute [r] model_servers
+        #   @return [::Array<::String>]
+        #     Output only. List of available model servers. Open-source model servers use
+        #     simplified, lowercase names (e.g., `vllm`).
+        # @!attribute [r] next_page_token
+        #   @return [::String]
+        #     Output only. A token which may be sent as
+        #     [page_token][FetchModelServersResponse.page_token] in a subsequent
+        #     `FetchModelServersResponse` call to retrieve the next page of results.
+        #     If this field is omitted or empty, then there are no more results to
+        #     return.
+        class FetchModelServersResponse
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Request message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_server_versions GkeInferenceQuickstart.FetchModelServerVersions}.
+        # @!attribute [rw] model
+        #   @return [::String]
+        #     Required. The model for which to list model server versions. Open-source
+        #     models follow the Huggingface Hub `owner/model_name` format. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_models GkeInferenceQuickstart.FetchModels}
+        #     to find available models.
+        # @!attribute [rw] model_server
+        #   @return [::String]
+        #     Required. The model server for which to list versions. Open-source model
+        #     servers use simplified, lowercase names (e.g., `vllm`). Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_servers GkeInferenceQuickstart.FetchModelServers}
+        #     to find available model servers.
+        # @!attribute [rw] page_size
+        #   @return [::Integer]
+        #     Optional. The target number of results to return in a single response.
+        #     If not specified, a default value will be chosen by the service.
+        #     Note that the response may include a partial list and a caller should
+        #     only rely on the response's
+        #     {::Google::Cloud::GkeRecommender::V1::FetchModelServerVersionsResponse#next_page_token next_page_token}
+        #     to determine if there are more instances left to be queried.
+        # @!attribute [rw] page_token
+        #   @return [::String]
+        #     Optional. The value of
+        #     {::Google::Cloud::GkeRecommender::V1::FetchModelServerVersionsResponse#next_page_token next_page_token}
+        #     received from a previous `FetchModelServerVersionsRequest` call.
+        #     Provide this to retrieve the subsequent page in a multi-page list of
+        #     results. When paginating, all other parameters provided to
+        #     `FetchModelServerVersionsRequest` must match the call that provided the
+        #     page token.
+        class FetchModelServerVersionsRequest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Response message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_server_versions GkeInferenceQuickstart.FetchModelServerVersions}.
+        # @!attribute [r] model_server_versions
+        #   @return [::Array<::String>]
+        #     Output only. A list of available model server versions.
+        # @!attribute [r] next_page_token
+        #   @return [::String]
+        #     Output only. A token which may be sent as
+        #     [page_token][FetchModelServerVersionsResponse.page_token] in a subsequent
+        #     `FetchModelServerVersionsResponse` call to retrieve the next page of
+        #     results. If this field is omitted or empty, then there are no more results
+        #     to return.
+        class FetchModelServerVersionsResponse
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Request message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_benchmarking_data GkeInferenceQuickstart.FetchBenchmarkingData}.
+        # @!attribute [rw] model_server_info
+        #   @return [::Google::Cloud::GkeRecommender::V1::ModelServerInfo]
+        #     Required. The model server configuration to get benchmarking data for. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}
+        #     to find valid configurations.
+        # @!attribute [rw] instance_type
+        #   @return [::String]
+        #     Optional. The instance type to filter benchmarking data. Instance types are
+        #     in the format `a2-highgpu-1g`. If not provided, all instance types for the
+        #     given profile's `model_server_info` will be returned. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}
+        #     to find available instance types.
+        # @!attribute [rw] pricing_model
+        #   @return [::String]
+        #     Optional. The pricing model to use for the benchmarking data. Defaults to
+        #     `spot`.
+        class FetchBenchmarkingDataRequest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Response message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_benchmarking_data GkeInferenceQuickstart.FetchBenchmarkingData}.
+        # @!attribute [r] profile
+        #   @return [::Array<::Google::Cloud::GkeRecommender::V1::Profile>]
+        #     Output only. List of profiles containing their respective benchmarking
+        #     data.
+        class FetchBenchmarkingDataResponse
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Request message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}.
+        # @!attribute [rw] model
+        #   @return [::String]
+        #     Optional. The model to filter profiles by. Open-source models follow the
+        #     Huggingface Hub `owner/model_name` format. If not provided, all models are
+        #     returned. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_models GkeInferenceQuickstart.FetchModels}
+        #     to find available models.
+        # @!attribute [rw] model_server
+        #   @return [::String]
+        #     Optional. The model server to filter profiles by. If not provided, all
+        #     model servers are returned. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_servers GkeInferenceQuickstart.FetchModelServers}
+        #     to find available model servers for a given model.
+        # @!attribute [rw] model_server_version
+        #   @return [::String]
+        #     Optional. The model server version to filter profiles by. If not provided,
+        #     all model server versions are returned. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_server_versions GkeInferenceQuickstart.FetchModelServerVersions}
+        #     to find available versions for a given model and server.
+        # @!attribute [rw] performance_requirements
+        #   @return [::Google::Cloud::GkeRecommender::V1::PerformanceRequirements]
+        #     Optional. The performance requirements to filter profiles. Profiles that do
+        #     not meet these requirements are filtered out. If not provided, all profiles
+        #     are returned.
+        # @!attribute [rw] page_size
+        #   @return [::Integer]
+        #     Optional. The target number of results to return in a single response. If
+        #     not specified, a default value will be chosen by the service. Note that the
+        #     response may include a partial list and a caller should only rely on the
+        #     response's
+        #     {::Google::Cloud::GkeRecommender::V1::FetchProfilesResponse#next_page_token next_page_token}
+        #     to determine if there are more instances left to be queried.
+        # @!attribute [rw] page_token
+        #   @return [::String]
+        #     Optional. The value of
+        #     {::Google::Cloud::GkeRecommender::V1::FetchProfilesResponse#next_page_token next_page_token}
+        #     received from a previous `FetchProfilesRequest` call.
+        #     Provide this to retrieve the subsequent page in a multi-page list of
+        #     results. When paginating, all other parameters provided to
+        #     `FetchProfilesRequest` must match the call that provided the page
+        #     token.
+        class FetchProfilesRequest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Performance requirements for a profile and or model deployment.
+        # @!attribute [rw] target_ntpot_milliseconds
+        #   @return [::Integer]
+        #     Optional. The target Normalized Time Per Output Token (NTPOT) in
+        #     milliseconds. NTPOT is calculated as `request_latency /
+        #     total_output_tokens`. If not provided, this target will not be enforced.
+        # @!attribute [rw] target_ttft_milliseconds
+        #   @return [::Integer]
+        #     Optional. The target Time To First Token (TTFT) in milliseconds. TTFT is
+        #     the time it takes to generate the first token for a request.  If not
+        #     provided, this target will not be enforced.
+        # @!attribute [rw] target_cost
+        #   @return [::Google::Cloud::GkeRecommender::V1::Cost]
+        #     Optional. The target cost for running a profile's model server. If not
+        #     provided, this requirement will not be enforced.
+        class PerformanceRequirements
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Represents an amount of money in a specific currency.
+        # @!attribute [r] units
+        #   @return [::Integer]
+        #     Output only. The whole units of the amount.
+        #     For example if `currencyCode` is `"USD"`, then 1 unit is one US dollar.
+        # @!attribute [r] nanos
+        #   @return [::Integer]
+        #     Output only. Number of nano (10^-9) units of the amount.
+        #     The value must be between -999,999,999 and +999,999,999 inclusive.
+        #     If `units` is positive, `nanos` must be positive or zero.
+        #     If `units` is zero, `nanos` can be positive, zero, or negative.
+        #     If `units` is negative, `nanos` must be negative or zero.
+        #     For example $-1.75 is represented as `units`=-1 and `nanos`=-750,000,000.
+        class Amount
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Cost for running a model deployment on a given instance type. Currently, only
+        # USD currency code is supported.
+        # @!attribute [rw] cost_per_million_output_tokens
+        #   @return [::Google::Cloud::GkeRecommender::V1::Amount]
+        #     Optional. The cost per million output tokens, calculated as:
+        #     $/output token = GPU $/s / (1/output-to-input-cost-ratio * input tokens/s +
+        #     output tokens/s)
+        # @!attribute [rw] cost_per_million_input_tokens
+        #   @return [::Google::Cloud::GkeRecommender::V1::Amount]
+        #     Optional. The cost per million input tokens. $/input token = ($/output
+        #     token) / output-to-input-cost-ratio.
+        # @!attribute [rw] pricing_model
+        #   @return [::String]
+        #     Optional. The pricing model used to calculate the cost. Can be one of:
+        #     `3-years-cud`, `1-year-cud`, `on-demand`, `spot`. If not provided, `spot`
+        #     will be used.
+        # @!attribute [rw] output_input_cost_ratio
+        #   @return [::Float]
+        #     Optional. The output-to-input cost ratio. This determines how the total GPU
+        #     cost is split between input and output tokens. If not provided, `4.0` is
+        #     used, assuming a 4:1 output:input cost ratio.
+        class Cost
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Represents a range of throughput values in tokens per second.
+        # @!attribute [r] min
+        #   @return [::Integer]
+        #     Output only. The minimum value of the range.
+        # @!attribute [r] max
+        #   @return [::Integer]
+        #     Output only. The maximum value of the range.
+        class TokensPerSecondRange
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Represents a range of latency values in milliseconds.
+        # @!attribute [r] min
+        #   @return [::Integer]
+        #     Output only. The minimum value of the range.
+        # @!attribute [r] max
+        #   @return [::Integer]
+        #     Output only. The maximum value of the range.
+        class MillisecondRange
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Performance range for a model deployment.
+        # @!attribute [r] throughput_output_range
+        #   @return [::Google::Cloud::GkeRecommender::V1::TokensPerSecondRange]
+        #     Output only. The range of throughput in output tokens per second. This is
+        #     measured as total_output_tokens_generated_by_server /
+        #     elapsed_time_in_seconds.
+        # @!attribute [r] ttft_range
+        #   @return [::Google::Cloud::GkeRecommender::V1::MillisecondRange]
+        #     Output only. The range of TTFT (Time To First Token) in milliseconds. TTFT
+        #     is the time it takes to generate the first token for a request.
+        # @!attribute [r] ntpot_range
+        #   @return [::Google::Cloud::GkeRecommender::V1::MillisecondRange]
+        #     Output only. The range of NTPOT (Normalized Time Per Output Token) in
+        #     milliseconds. NTPOT is the request latency normalized by the number of
+        #     output tokens, measured as request_latency / total_output_tokens.
+        class PerformanceRange
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Response message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}.
+        # @!attribute [r] profile
+        #   @return [::Array<::Google::Cloud::GkeRecommender::V1::Profile>]
+        #     Output only. List of profiles that match the given model server info and
+        #     performance requirements (if provided).
+        # @!attribute [r] performance_range
+        #   @return [::Google::Cloud::GkeRecommender::V1::PerformanceRange]
+        #     Output only. The combined range of performance values observed across all
+        #     profiles in this response.
+        # @!attribute [r] comments
+        #   @return [::String]
+        #     Output only. Additional comments related to the response.
+        # @!attribute [r] next_page_token
+        #   @return [::String]
+        #     Output only. A token which may be sent as
+        #     [page_token][FetchProfilesResponse.page_token] in a subsequent
+        #     `FetchProfilesResponse` call to retrieve the next page of results. If this
+        #     field is omitted or empty, then there are no more results to return.
+        class FetchProfilesResponse
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Model server information gives. Valid model server info combinations can
+        # be found using
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}.
+        # @!attribute [rw] model
+        #   @return [::String]
+        #     Required. The model. Open-source models follow the Huggingface Hub
+        #     `owner/model_name` format. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_models GkeInferenceQuickstart.FetchModels}
+        #     to find available models.
+        # @!attribute [rw] model_server
+        #   @return [::String]
+        #     Required. The model server. Open-source model servers use simplified,
+        #     lowercase names (e.g., `vllm`). Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_servers GkeInferenceQuickstart.FetchModelServers}
+        #     to find available servers.
+        # @!attribute [rw] model_server_version
+        #   @return [::String]
+        #     Optional. The model server version. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_model_server_versions GkeInferenceQuickstart.FetchModelServerVersions}
+        #     to find available versions. If not provided, the latest available version
+        #     is used.
+        class ModelServerInfo
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Resources used by a model deployment.
+        # @!attribute [r] accelerator_count
+        #   @return [::Integer]
+        #     Output only. The number of accelerators (e.g., GPUs or TPUs) used by the
+        #     model deployment on the Kubernetes node.
+        class ResourcesUsed
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Performance statistics for a model deployment.
+        # @!attribute [r] queries_per_second
+        #   @return [::Float]
+        #     Output only. The number of queries per second.
+        #     Note: This metric can vary widely based on context length and may not be a
+        #     reliable measure of LLM throughput.
+        # @!attribute [r] output_tokens_per_second
+        #   @return [::Integer]
+        #     Output only. The number of output tokens per second. This is the throughput
+        #     measured as total_output_tokens_generated_by_server /
+        #     elapsed_time_in_seconds.
+        # @!attribute [r] ntpot_milliseconds
+        #   @return [::Integer]
+        #     Output only. The Normalized Time Per Output Token (NTPOT) in milliseconds.
+        #     This is the request latency normalized by the number of output tokens,
+        #     measured as request_latency / total_output_tokens.
+        # @!attribute [r] ttft_milliseconds
+        #   @return [::Integer]
+        #     Output only. The Time To First Token (TTFT) in milliseconds. This is the
+        #     time it takes to generate the first token for a request.
+        # @!attribute [r] cost
+        #   @return [::Array<::Google::Cloud::GkeRecommender::V1::Cost>]
+        #     Output only. The cost of running the model deployment.
+        class PerformanceStats
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # A profile containing information about a model deployment.
+        # @!attribute [r] model_server_info
+        #   @return [::Google::Cloud::GkeRecommender::V1::ModelServerInfo]
+        #     Output only. The model server configuration. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}
+        #     to find valid configurations.
+        # @!attribute [r] accelerator_type
+        #   @return [::String]
+        #     Output only. The accelerator type. Expected format: `nvidia-h100-80gb`.
+        # @!attribute [r] tpu_topology
+        #   @return [::String]
+        #     Output only. The TPU topology (if applicable).
+        # @!attribute [r] instance_type
+        #   @return [::String]
+        #     Output only. The instance type. Expected format: `a2-highgpu-1g`.
+        # @!attribute [r] resources_used
+        #   @return [::Google::Cloud::GkeRecommender::V1::ResourcesUsed]
+        #     Output only. The resources used by the model deployment.
+        # @!attribute [r] performance_stats
+        #   @return [::Array<::Google::Cloud::GkeRecommender::V1::PerformanceStats>]
+        #     Output only. The performance statistics for this profile.
+        class Profile
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Request message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#generate_optimized_manifest GkeInferenceQuickstart.GenerateOptimizedManifest}.
+        # @!attribute [rw] model_server_info
+        #   @return [::Google::Cloud::GkeRecommender::V1::ModelServerInfo]
+        #     Required. The model server configuration to generate the manifest for. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}
+        #     to find valid configurations.
+        # @!attribute [rw] accelerator_type
+        #   @return [::String]
+        #     Required. The accelerator type. Use
+        #     {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#fetch_profiles GkeInferenceQuickstart.FetchProfiles}
+        #     to find valid accelerators for a given `model_server_info`.
+        # @!attribute [rw] kubernetes_namespace
+        #   @return [::String]
+        #     Optional. The kubernetes namespace to deploy the manifests in.
+        # @!attribute [rw] performance_requirements
+        #   @return [::Google::Cloud::GkeRecommender::V1::PerformanceRequirements]
+        #     Optional. The performance requirements to use for generating Horizontal Pod
+        #     Autoscaler (HPA) resources. If provided, the manifest includes HPA
+        #     resources to adjust the model server replica count to maintain the
+        #     specified targets (e.g., NTPOT, TTFT) at a P50 latency. Cost targets are
+        #     not currently supported for HPA generation. If the specified targets are
+        #     not achievable, the HPA manifest will not be generated.
+        # @!attribute [rw] storage_config
+        #   @return [::Google::Cloud::GkeRecommender::V1::StorageConfig]
+        #     Optional. The storage configuration for the model. If not provided, the
+        #     model is loaded from Huggingface.
+        class GenerateOptimizedManifestRequest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # A Kubernetes manifest.
+        # @!attribute [r] kind
+        #   @return [::String]
+        #     Output only. Kubernetes resource kind.
+        # @!attribute [r] api_version
+        #   @return [::String]
+        #     Output only. Kubernetes API version.
+        # @!attribute [r] content
+        #   @return [::String]
+        #     Output only. YAML content.
+        class KubernetesManifest
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Response message for
+        # {::Google::Cloud::GkeRecommender::V1::GkeInferenceQuickstart::Client#generate_optimized_manifest GkeInferenceQuickstart.GenerateOptimizedManifest}.
+        # @!attribute [r] kubernetes_manifests
+        #   @return [::Array<::Google::Cloud::GkeRecommender::V1::KubernetesManifest>]
+        #     Output only. A list of generated Kubernetes manifests.
+        # @!attribute [r] comments
+        #   @return [::Array<::String>]
+        #     Output only. Comments related to deploying the generated manifests.
+        # @!attribute [r] manifest_version
+        #   @return [::String]
+        #     Output only. Additional information about the versioned dependencies used
+        #     to generate the manifests. See [Run best practice inference with GKE
+        #     Inference Quickstart
+        #     recipes](https://cloud.google.com/kubernetes-engine/docs/how-to/machine-learning/inference/inference-quickstart)
+        #     for details.
+        class GenerateOptimizedManifestResponse
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+
+        # Storage configuration for a model deployment.
+        # @!attribute [rw] model_bucket_uri
+        #   @return [::String]
+        #     Optional. The Google Cloud Storage bucket URI to load the model from. This
+        #     URI must point to the directory containing the model's config file
+        #     (`config.json`) and model weights. A tuned GCSFuse setup can improve
+        #     LLM Pod startup time by more than 7x. Expected format:
+        #     `gs://<bucket-name>/<path-to-model>`.
+        # @!attribute [rw] xla_cache_bucket_uri
+        #   @return [::String]
+        #     Optional. The URI for the GCS bucket containing the XLA compilation cache.
+        #     If using TPUs, the XLA cache will be written to the same path as
+        #     `model_bucket_uri`. This can speed up vLLM model preparation for repeated
+        #     deployments.
+        class StorageConfig
+          include ::Google::Protobuf::MessageExts
+          extend ::Google::Protobuf::MessageExts::ClassMethods
+        end
+      end
+    end
+  end
+end

data/proto_docs/google/protobuf/duration.rb

@@ -0,0 +1,98 @@
+# frozen_string_literal: true
+
+# Copyright 2025 Google LLC
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     https://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# Auto-generated by gapic-generator-ruby. DO NOT EDIT!
+
+
+module Google
+  module Protobuf
+    # A Duration represents a signed, fixed-length span of time represented
+    # as a count of seconds and fractions of seconds at nanosecond
+    # resolution. It is independent of any calendar and concepts like "day"
+    # or "month". It is related to Timestamp in that the difference between
+    # two Timestamp values is a Duration and it can be added or subtracted
+    # from a Timestamp. Range is approximately +-10,000 years.
+    #
+    # # Examples
+    #
+    # Example 1: Compute Duration from two Timestamps in pseudo code.
+    #
+    #     Timestamp start = ...;
+    #     Timestamp end = ...;
+    #     Duration duration = ...;
+    #
+    #     duration.seconds = end.seconds - start.seconds;
+    #     duration.nanos = end.nanos - start.nanos;
+    #
+    #     if (duration.seconds < 0 && duration.nanos > 0) {
+    #       duration.seconds += 1;
+    #       duration.nanos -= 1000000000;
+    #     } else if (duration.seconds > 0 && duration.nanos < 0) {
+    #       duration.seconds -= 1;
+    #       duration.nanos += 1000000000;
+    #     }
+    #
+    # Example 2: Compute Timestamp from Timestamp + Duration in pseudo code.
+    #
+    #     Timestamp start = ...;
+    #     Duration duration = ...;
+    #     Timestamp end = ...;
+    #
+    #     end.seconds = start.seconds + duration.seconds;
+    #     end.nanos = start.nanos + duration.nanos;
+    #
+    #     if (end.nanos < 0) {
+    #       end.seconds -= 1;
+    #       end.nanos += 1000000000;
+    #     } else if (end.nanos >= 1000000000) {
+    #       end.seconds += 1;
+    #       end.nanos -= 1000000000;
+    #     }
+    #
+    # Example 3: Compute Duration from datetime.timedelta in Python.
+    #
+    #     td = datetime.timedelta(days=3, minutes=10)
+    #     duration = Duration()
+    #     duration.FromTimedelta(td)
+    #
+    # # JSON Mapping
+    #
+    # In JSON format, the Duration type is encoded as a string rather than an
+    # object, where the string ends in the suffix "s" (indicating seconds) and
+    # is preceded by the number of seconds, with nanoseconds expressed as
+    # fractional seconds. For example, 3 seconds with 0 nanoseconds should be
+    # encoded in JSON format as "3s", while 3 seconds and 1 nanosecond should
+    # be expressed in JSON format as "3.000000001s", and 3 seconds and 1
+    # microsecond should be expressed in JSON format as "3.000001s".
+    # @!attribute [rw] seconds
+    #   @return [::Integer]
+    #     Signed seconds of the span of time. Must be from -315,576,000,000
+    #     to +315,576,000,000 inclusive. Note: these bounds are computed from:
+    #     60 sec/min * 60 min/hr * 24 hr/day * 365.25 days/year * 10000 years
+    # @!attribute [rw] nanos
+    #   @return [::Integer]
+    #     Signed fractions of a second at nanosecond resolution of the span
+    #     of time. Durations less than one second are represented with a 0
+    #     `seconds` field and a positive or negative `nanos` field. For durations
+    #     of one second or more, a non-zero value for the `nanos` field must be
+    #     of the same sign as the `seconds` field. Must be from -999,999,999
+    #     to +999,999,999 inclusive.
+    class Duration
+      include ::Google::Protobuf::MessageExts
+      extend ::Google::Protobuf::MessageExts::ClassMethods
+    end
+  end
+end