@daocloud-proto/baize 0.103.3 → 0.104.1

package/common/k8s.pb.ts

@@ -25,4 +25,78 @@ export type KubeVolume = {
   name?: string
   mountPath?: string
   readOnly?: boolean
+}
+
+export type PodConfig = {
+  kubeEnvs?: KubeEnv[]
+  kubeVolumes?: KubeVolume[]
+  resources?: Resources
+  affinity?: Affinity
+  schedulerName?: string
+  priorityClass?: string
+  queue?: string
+}
+
+export type Affinity = {
+  nodeAffinity?: NodeAffinity
+  podAffinity?: PodAffinity
+  podAntiAffinity?: PodAntiAffinity
+}
+
+export type PodAntiAffinity = {
+  requiredDuringSchedulingIgnoredDuringExecution?: PodAffinityTerm[]
+  preferredDuringSchedulingIgnoredDuringExecution?: WeightedPodAffinityTerm[]
+}
+
+export type PodAffinityTerm = {
+  labelSelector?: LabelSelector
+  namespaces?: string[]
+  topologyKey?: string
+  namespaceSelector?: LabelSelector
+}
+
+export type LabelSelector = {
+  matchLabels?: {[key: string]: string}
+  matchExpressions?: LabelSelectorRequirement[]
+}
+
+export type WeightedPodAffinityTerm = {
+  weight?: number
+  podAffinityTerm?: PodAffinityTerm
+}
+
+export type PodAffinity = {
+  requiredDuringSchedulingIgnoredDuringExecution?: PodAffinityTerm[]
+  preferredDuringSchedulingIgnoredDuringExecution?: WeightedPodAffinityTerm[]
+}
+
+export type NodeAffinity = {
+  requiredDuringSchedulingIgnoredDuringExecution?: NodeSelector
+  preferredDuringSchedulingIgnoredDuringExecution?: PreferredSchedulingTerm[]
+}
+
+export type NodeSelector = {
+  nodeSelectorTerms?: NodeSelectorTerm[]
+}
+
+export type PreferredSchedulingTerm = {
+  weight?: number
+  preference?: NodeSelectorTerm
+}
+
+export type NodeSelectorTerm = {
+  matchExpressions?: NodeSelectorRequirement[]
+  matchFields?: NodeSelectorRequirement[]
+}
+
+export type NodeSelectorRequirement = {
+  key?: string
+  operator?: string
+  values?: string[]
+}
+
+export type LabelSelectorRequirement = {
+  key?: string
+  operator?: string
+  values?: string[]
 }

package/management-api/job/v1alpha1/common.pb.ts

@@ -6,6 +6,15 @@
 
 import * as BaizeCommonK8s from "../../../common/k8s.pb"
 
+type Absent<T, K extends keyof T> = { [k in Exclude<keyof T, K>]?: undefined };
+type OneOf<T> =
+  | { [k in keyof T]?: undefined }
+  | (
+    keyof T extends infer K ?
+      (K extends string & keyof T ? { [k in K]: T[K] } & Absent<T, K>
+        : never)
+    : never);
+
 export enum QueueType {
   QUEUE_TYPE_UNSPECIFIED = "QUEUE_TYPE_UNSPECIFIED",
   KUEUE = "KUEUE",
@@ -32,7 +41,8 @@ export type CommonConfig = {
   description?: string
 }
 
-export type JobCreationBaseConfig = {
+
+/* baize modified */ export type BaseJobCreationBaseConfig = {
   image?: string
   imagePullSecret?: string
   command?: string[]
@@ -41,8 +51,12 @@ export type JobCreationBaseConfig = {
   annotations?: {[key: string]: string}
   kubeEnvs?: BaizeCommonK8s.KubeEnv[]
   kubeVolumes?: BaizeCommonK8s.KubeVolume[]
+  affinity?: BaizeCommonK8s.Affinity
 }
 
+export type JobCreationBaseConfig = BaseJobCreationBaseConfig
+  & OneOf<{ tolerationSeconds: string }>
+
 export type JobRoleDifferenceConfig = {
   replicas?: number
   resources?: BaizeCommonK8s.Resources

package/management-api/job/v1alpha1/job.pb.ts

@@ -39,10 +39,12 @@ export enum JobActionRequestAction {
   priorityClass?: string
   runningDuration?: number
   totalResources?: BaizeCommonK8s.Resources
+  affinity?: BaizeCommonK8s.Affinity
 }
 
 export type Job = BaseJob
   & OneOf<{ pytorch: BaizeManagement_apiJobV1alpha1Pytorch.PyTorchJob; tensorflow: BaizeManagement_apiJobV1alpha1Tfjob.TFJob; paddle: BaizeManagement_apiJobV1alpha1Paddle.PaddleJob }>
+  & OneOf<{ tolerationSeconds: string }>
 
 export type ListJobsRequest = {
   workspace?: number

package/management-api/pod/v1alpha1/pod.pb.ts

@@ -49,7 +49,7 @@ export type PodInstanceListResponse = {
   page?: BaizeCommonCommon.Pagination
 }
 
-export class podsManagement {
+export class PodsManagement {
   static GetPodInstanceList(req: PodRequest, initReq?: fm.InitReq): Promise<PodInstanceListResponse> {
     return fm.fetchReq<PodRequest, PodInstanceListResponse>(`/apis/baize.io/v1alpha1/workspaces/${req["workspace"]}/clusters/${req["cluster"]}/namespaces/${req["namespace"]}/resources/${req["name"]}/instances?${fm.renderURLSearchParams(req, ["workspace", "cluster", "namespace", "name"])}`, {...initReq, method: "GET"})
   }

package/management-api/queue/v1alpha1/queue.pb.ts

@@ -60,6 +60,10 @@ export type ListQueueResponse = {
   page?: BaizeCommonCommon.Pagination
 }
 
+export type QueueJSON = {
+  data?: string
+}
+
 export type CreateQueueRequest = {
   type?: QueueType
   cluster?: string
@@ -139,6 +143,9 @@ export class QueueManagement {
   static GetQueue(req: GetQueueRequest, initReq?: fm.InitReq): Promise<Queue> {
     return fm.fetchReq<GetQueueRequest, Queue>(`/apis/baize.io/v1alpha1/clusters/${req["cluster"]}/queues/${req["name"]}?${fm.renderURLSearchParams(req, ["cluster", "name"])}`, {...initReq, method: "GET"})
   }
+  static GetQueueByJSON(req: GetQueueRequest, initReq?: fm.InitReq): Promise<QueueJSON> {
+    return fm.fetchReq<GetQueueRequest, QueueJSON>(`/apis/baize.io/v1alpha1/clusters/${req["cluster"]}/queues/${req["name"]}/json?${fm.renderURLSearchParams(req, ["cluster", "name"])}`, {...initReq, method: "GET"})
+  }
   static DeleteQueue(req: DeleteQueueRequest, initReq?: fm.InitReq): Promise<Queue> {
     return fm.fetchReq<DeleteQueueRequest, Queue>(`/apis/baize.io/v1alpha1/clusters/${req["cluster"]}/queues/${req["name"]}`, {...initReq, method: "DELETE", body: JSON.stringify(req, fm.replacer)})
   }

package/management-api/serving/triton/model_config.pb.ts

@@ -0,0 +1,368 @@
+/* eslint-disable */
+// @ts-nocheck
+/*
+* This file is a generated Typescript file for GRPC Gateway, DO NOT MODIFY
+*/
+
+type Absent<T, K extends keyof T> = { [k in Exclude<keyof T, K>]?: undefined };
+type OneOf<T> =
+  | { [k in keyof T]?: undefined }
+  | (
+    keyof T extends infer K ?
+      (K extends string & keyof T ? { [k in K]: T[K] } & Absent<T, K>
+        : never)
+    : never);
+
+export enum DataType {
+  TYPE_INVALID = "TYPE_INVALID",
+  TYPE_BOOL = "TYPE_BOOL",
+  TYPE_UINT8 = "TYPE_UINT8",
+  TYPE_UINT16 = "TYPE_UINT16",
+  TYPE_UINT32 = "TYPE_UINT32",
+  TYPE_UINT64 = "TYPE_UINT64",
+  TYPE_INT8 = "TYPE_INT8",
+  TYPE_INT16 = "TYPE_INT16",
+  TYPE_INT32 = "TYPE_INT32",
+  TYPE_INT64 = "TYPE_INT64",
+  TYPE_FP16 = "TYPE_FP16",
+  TYPE_FP32 = "TYPE_FP32",
+  TYPE_FP64 = "TYPE_FP64",
+  TYPE_STRING = "TYPE_STRING",
+  TYPE_BF16 = "TYPE_BF16",
+}
+
+export enum ModelInstanceGroupKind {
+  KIND_AUTO = "KIND_AUTO",
+  KIND_GPU = "KIND_GPU",
+  KIND_CPU = "KIND_CPU",
+  KIND_MODEL = "KIND_MODEL",
+}
+
+export enum ModelInstanceGroupSecondaryDeviceSecondaryDeviceKind {
+  KIND_NVDLA = "KIND_NVDLA",
+}
+
+export enum ModelInputFormat {
+  FORMAT_NONE = "FORMAT_NONE",
+  FORMAT_NHWC = "FORMAT_NHWC",
+  FORMAT_NCHW = "FORMAT_NCHW",
+}
+
+export enum BatchInputKind {
+  BATCH_ELEMENT_COUNT = "BATCH_ELEMENT_COUNT",
+  BATCH_ACCUMULATED_ELEMENT_COUNT = "BATCH_ACCUMULATED_ELEMENT_COUNT",
+  BATCH_ACCUMULATED_ELEMENT_COUNT_WITH_ZERO = "BATCH_ACCUMULATED_ELEMENT_COUNT_WITH_ZERO",
+  BATCH_MAX_ELEMENT_COUNT_AS_SHAPE = "BATCH_MAX_ELEMENT_COUNT_AS_SHAPE",
+  BATCH_ITEM_SHAPE = "BATCH_ITEM_SHAPE",
+  BATCH_ITEM_SHAPE_FLATTEN = "BATCH_ITEM_SHAPE_FLATTEN",
+}
+
+export enum BatchOutputKind {
+  BATCH_SCATTER_WITH_INPUT_SHAPE = "BATCH_SCATTER_WITH_INPUT_SHAPE",
+}
+
+export enum ModelOptimizationPolicyModelPriority {
+  PRIORITY_DEFAULT = "PRIORITY_DEFAULT",
+  PRIORITY_MAX = "PRIORITY_MAX",
+  PRIORITY_MIN = "PRIORITY_MIN",
+}
+
+export enum ModelQueuePolicyTimeoutAction {
+  REJECT = "REJECT",
+  DELAY = "DELAY",
+}
+
+export enum ModelSequenceBatchingControlKind {
+  CONTROL_SEQUENCE_START = "CONTROL_SEQUENCE_START",
+  CONTROL_SEQUENCE_READY = "CONTROL_SEQUENCE_READY",
+  CONTROL_SEQUENCE_END = "CONTROL_SEQUENCE_END",
+  CONTROL_SEQUENCE_CORRID = "CONTROL_SEQUENCE_CORRID",
+}
+
+export type ModelRateLimiterResource = {
+  name?: string
+  global?: boolean
+  count?: number
+}
+
+export type ModelRateLimiter = {
+  resources?: ModelRateLimiterResource[]
+  priority?: number
+}
+
+export type ModelInstanceGroupSecondaryDevice = {
+  kind?: ModelInstanceGroupSecondaryDeviceSecondaryDeviceKind
+  deviceId?: string
+}
+
+export type ModelInstanceGroup = {
+  name?: string
+  kind?: ModelInstanceGroupKind
+  count?: number
+  rateLimiter?: ModelRateLimiter
+  gpus?: number[]
+  secondaryDevices?: ModelInstanceGroupSecondaryDevice[]
+  profile?: string[]
+  passive?: boolean
+  hostPolicy?: string
+}
+
+export type ModelTensorReshape = {
+  shape?: string[]
+}
+
+export type ModelInput = {
+  name?: string
+  dataType?: DataType
+  format?: ModelInputFormat
+  dims?: string[]
+  reshape?: ModelTensorReshape
+  isShapeTensor?: boolean
+  allowRaggedBatch?: boolean
+  optional?: boolean
+}
+
+export type ModelOutput = {
+  name?: string
+  dataType?: DataType
+  dims?: string[]
+  reshape?: ModelTensorReshape
+  labelFilename?: string
+  isShapeTensor?: boolean
+}
+
+export type BatchInput = {
+  kind?: BatchInputKind
+  targetName?: string[]
+  dataType?: DataType
+  sourceInput?: string[]
+}
+
+export type BatchOutput = {
+  targetName?: string[]
+  kind?: BatchOutputKind
+  sourceInput?: string[]
+}
+
+export type ModelVersionPolicyLatest = {
+  numVersions?: number
+}
+
+export type ModelVersionPolicyAll = {
+}
+
+export type ModelVersionPolicySpecific = {
+  versions?: string[]
+}
+
+
+/* baize modified */ export type BaseModelVersionPolicy = {
+}
+
+export type ModelVersionPolicy = BaseModelVersionPolicy
+  & OneOf<{ latest: ModelVersionPolicyLatest; all: ModelVersionPolicyAll; specific: ModelVersionPolicySpecific }>
+
+export type ModelOptimizationPolicyGraph = {
+  level?: number
+}
+
+export type ModelOptimizationPolicyCudaGraphSpecShape = {
+  dim?: string[]
+}
+
+export type ModelOptimizationPolicyCudaGraphSpecLowerBound = {
+  batchSize?: number
+  input?: {[key: string]: ModelOptimizationPolicyCudaGraphSpecShape}
+}
+
+export type ModelOptimizationPolicyCudaGraphSpec = {
+  batchSize?: number
+  input?: {[key: string]: ModelOptimizationPolicyCudaGraphSpecShape}
+  graphLowerBound?: ModelOptimizationPolicyCudaGraphSpecLowerBound
+}
+
+export type ModelOptimizationPolicyCuda = {
+  graphs?: boolean
+  busyWaitEvents?: boolean
+  graphSpec?: ModelOptimizationPolicyCudaGraphSpec[]
+  outputCopyStream?: boolean
+}
+
+export type ModelOptimizationPolicyExecutionAcceleratorsAccelerator = {
+  name?: string
+  parameters?: {[key: string]: string}
+}
+
+export type ModelOptimizationPolicyExecutionAccelerators = {
+  gpuExecutionAccelerator?: ModelOptimizationPolicyExecutionAcceleratorsAccelerator[]
+  cpuExecutionAccelerator?: ModelOptimizationPolicyExecutionAcceleratorsAccelerator[]
+}
+
+export type ModelOptimizationPolicyPinnedMemoryBuffer = {
+  enable?: boolean
+}
+
+export type ModelOptimizationPolicy = {
+  graph?: ModelOptimizationPolicyGraph
+  priority?: ModelOptimizationPolicyModelPriority
+  cuda?: ModelOptimizationPolicyCuda
+  executionAccelerators?: ModelOptimizationPolicyExecutionAccelerators
+  inputPinnedMemory?: ModelOptimizationPolicyPinnedMemoryBuffer
+  outputPinnedMemory?: ModelOptimizationPolicyPinnedMemoryBuffer
+  gatherKernelBufferThreshold?: number
+  eagerBatching?: boolean
+}
+
+export type ModelQueuePolicy = {
+  timeoutAction?: ModelQueuePolicyTimeoutAction
+  defaultTimeoutMicroseconds?: string
+  allowTimeoutOverride?: boolean
+  maxQueueSize?: number
+}
+
+export type ModelDynamicBatching = {
+  preferredBatchSize?: number[]
+  maxQueueDelayMicroseconds?: string
+  preserveOrdering?: boolean
+  priorityLevels?: string
+  defaultPriorityLevel?: string
+  defaultQueuePolicy?: ModelQueuePolicy
+  priorityQueuePolicy?: {[key: string]: ModelQueuePolicy}
+}
+
+export type ModelSequenceBatchingControl = {
+  kind?: ModelSequenceBatchingControlKind
+  int32FalseTrue?: number[]
+  fp32FalseTrue?: number[]
+  boolFalseTrue?: boolean[]
+  dataType?: DataType
+}
+
+export type ModelSequenceBatchingControlInput = {
+  name?: string
+  control?: ModelSequenceBatchingControl[]
+}
+
+
+/* baize modified */ export type BaseModelSequenceBatchingInitialState = {
+  dataType?: DataType
+  dims?: string[]
+  name?: string
+}
+
+export type ModelSequenceBatchingInitialState = BaseModelSequenceBatchingInitialState
+  & OneOf<{ zeroData: boolean; dataFile: string }>
+
+export type ModelSequenceBatchingState = {
+  inputName?: string
+  outputName?: string
+  dataType?: DataType
+  dims?: string[]
+  initialState?: ModelSequenceBatchingInitialState[]
+  useSameBufferForInputOutput?: boolean
+  useGrowableMemory?: boolean
+}
+
+export type ModelSequenceBatchingStrategyDirect = {
+  maxQueueDelayMicroseconds?: string
+  minimumSlotUtilization?: number
+}
+
+export type ModelSequenceBatchingStrategyOldest = {
+  maxCandidateSequences?: number
+  preferredBatchSize?: number[]
+  maxQueueDelayMicroseconds?: string
+  preserveOrdering?: boolean
+}
+
+
+/* baize modified */ export type BaseModelSequenceBatching = {
+  maxSequenceIdleMicroseconds?: string
+  controlInput?: ModelSequenceBatchingControlInput[]
+  state?: ModelSequenceBatchingState[]
+  iterativeSequence?: boolean
+}
+
+export type ModelSequenceBatching = BaseModelSequenceBatching
+  & OneOf<{ direct: ModelSequenceBatchingStrategyDirect; oldest: ModelSequenceBatchingStrategyOldest }>
+
+export type ModelEnsemblingStep = {
+  modelName?: string
+  modelVersion?: string
+  inputMap?: {[key: string]: string}
+  outputMap?: {[key: string]: string}
+  modelNamespace?: string
+}
+
+export type ModelEnsembling = {
+  step?: ModelEnsemblingStep[]
+}
+
+export type ModelParameter = {
+  stringValue?: string
+}
+
+
+/* baize modified */ export type BaseModelWarmupInput = {
+  dataType?: DataType
+  dims?: string[]
+}
+
+export type ModelWarmupInput = BaseModelWarmupInput
+  & OneOf<{ zeroData: boolean; randomData: boolean; inputDataFile: string }>
+
+export type ModelWarmup = {
+  name?: string
+  batchSize?: number
+  inputs?: {[key: string]: ModelWarmupInput}
+  count?: number
+}
+
+export type ModelOperations = {
+  opLibraryFilename?: string[]
+}
+
+export type ModelTransactionPolicy = {
+  decoupled?: boolean
+}
+
+export type ModelRepositoryAgentsAgent = {
+  name?: string
+  parameters?: {[key: string]: string}
+}
+
+export type ModelRepositoryAgents = {
+  agents?: ModelRepositoryAgentsAgent[]
+}
+
+export type ModelResponseCache = {
+  enable?: boolean
+}
+
+
+/* baize modified */ export type BaseModelConfig = {
+  name?: string
+  platform?: string
+  backend?: string
+  runtime?: string
+  versionPolicy?: ModelVersionPolicy
+  maxBatchSize?: number
+  input?: ModelInput[]
+  output?: ModelOutput[]
+  batchInput?: BatchInput[]
+  batchOutput?: BatchOutput[]
+  optimization?: ModelOptimizationPolicy
+  instanceGroup?: ModelInstanceGroup[]
+  defaultModelFilename?: string
+  ccModelFilenames?: {[key: string]: string}
+  metricTags?: {[key: string]: string}
+  parameters?: {[key: string]: ModelParameter}
+  modelWarmup?: ModelWarmup[]
+  modelOperations?: ModelOperations
+  modelTransactionPolicy?: ModelTransactionPolicy
+  modelRepositoryAgents?: ModelRepositoryAgents
+  responseCache?: ModelResponseCache
+}
+
+export type ModelConfig = BaseModelConfig
+  & OneOf<{ dynamicBatching: ModelDynamicBatching; sequenceBatching: ModelSequenceBatching; ensembleScheduling: ModelEnsembling }>

package/management-api/serving/v1alpha1/serving.pb.ts

@@ -0,0 +1,212 @@
+/* eslint-disable */
+// @ts-nocheck
+/*
+* This file is a generated Typescript file for GRPC Gateway, DO NOT MODIFY
+*/
+
+import * as BaizeCommonCommon from "../../../common/common.pb"
+import * as BaizeCommonK8s from "../../../common/k8s.pb"
+import * as fm from "../../../fetch.pb"
+import * as GoogleProtobufTimestamp from "../../../google/protobuf/timestamp.pb"
+import * as BaizeManagement_apiServingTritonModel_config from "../triton/model_config.pb"
+
+type Absent<T, K extends keyof T> = { [k in Exclude<keyof T, K>]?: undefined };
+type OneOf<T> =
+  | { [k in keyof T]?: undefined }
+  | (
+    keyof T extends infer K ?
+      (K extends string & keyof T ? { [k in K]: T[K] } & Absent<T, K>
+        : never)
+    : never);
+
+export enum ServiceType {
+  SERVICE_TYPE_UNSPECIFIED = "SERVICE_TYPE_UNSPECIFIED",
+  NODE_PORT = "NODE_PORT",
+  LOAD_BALANCER = "LOAD_BALANCER",
+  CLUSTER_IP = "CLUSTER_IP",
+}
+
+export enum FrameworkType {
+  FRAMEWORK_TYPE_UNSPECIFIED = "FRAMEWORK_TYPE_UNSPECIFIED",
+  FRAMEWORK_TYPE_TRITON = "FRAMEWORK_TYPE_TRITON",
+}
+
+export enum FrameworkTritonBackend {
+  TRITON_BACKEND_UNSPECIFIED = "TRITON_BACKEND_UNSPECIFIED",
+  TRITON_BACKEND_PYTORCH = "TRITON_BACKEND_PYTORCH",
+  TRITON_BACKEND_TENSORFLOW = "TRITON_BACKEND_TENSORFLOW",
+  TRITON_BACKEND_VLLM = "TRITON_BACKEND_VLLM",
+  TRITON_BACKEND_ONNX = "TRITON_BACKEND_ONNX",
+}
+
+export enum InferenceServingStatusPhase {
+  PHASE_UNSPECIFIED = "PHASE_UNSPECIFIED",
+  PENDING = "PENDING",
+  UPDATING_OR_CREATING = "UPDATING_OR_CREATING",
+  RUNNING = "RUNNING",
+  FAILED = "FAILED",
+  DELETING = "DELETING",
+  STOPPED = "STOPPED",
+}
+
+export enum ServingAuthAuthType {
+  AUTH_TYPE_UNSPECIFIED = "AUTH_TYPE_UNSPECIFIED",
+  TRITON_RESTRICTED_KEY = "TRITON_RESTRICTED_KEY",
+}
+
+export type FrameworkTriton = {
+  backend?: FrameworkTritonBackend
+}
+
+
+/* baize modified */ export type BaseFramework = {
+  type?: FrameworkType
+}
+
+export type Framework = BaseFramework
+  & OneOf<{ triton: FrameworkTriton }>
+
+export type Model = {
+  name?: string
+  version?: string
+  modelPath?: string
+}
+
+export type ServingConfigVLLM = {
+  trustRemoteCode?: boolean
+  tensorParallelSize?: number
+}
+
+export type ServingConfigTritonModelConfig = {
+  inputs?: BaizeManagement_apiServingTritonModel_config.ModelInput[]
+  outputs?: BaizeManagement_apiServingTritonModel_config.ModelOutput[]
+  maxBatchSize?: number
+  customConfig?: string
+}
+
+
+/* baize modified */ export type BaseServingConfigTritonServingConfig = {
+  customModelConfig?: {[key: string]: string}
+}
+
+export type ServingConfigTritonServingConfig = BaseServingConfigTritonServingConfig
+  & OneOf<{ config: ServingConfigTritonModelConfig; vllm: ServingConfigVLLM }>
+
+
+/* baize modified */ export type BaseServingConfig = {
+  name?: string
+  version?: string
+  modelPath?: string
+}
+
+export type ServingConfig = BaseServingConfig
+  & OneOf<{ triton: ServingConfigTritonServingConfig }>
+
+export type ServiceConfig = {
+  serviceType?: ServiceType
+}
+
+export type InferenceServing = {
+  name?: string
+  cluster?: string
+  namespace?: string
+  models?: ServingConfig[]
+  serviceConfig?: ServiceConfig
+  replicas?: number
+  labels?: {[key: string]: string}
+  annotations?: {[key: string]: string}
+  framework?: Framework
+  podConfig?: BaizeCommonK8s.PodConfig
+  status?: InferenceServingStatus
+  lastUpdated?: GoogleProtobufTimestamp.Timestamp
+}
+
+export type InferenceServingStatusModelStatus = {
+  name?: string
+  accessPath?: string
+}
+
+export type InferenceServingStatus = {
+  phase?: InferenceServingStatusPhase
+  availableReplicas?: number
+  models?: InferenceServingStatusModelStatus[]
+  accessBaseUrl?: string
+}
+
+export type ServingAuthTritonRestrictedKeyValue = {
+  key?: string
+  value?: string
+}
+
+
+/* baize modified */ export type BaseServingAuth = {
+  authType?: ServingAuthAuthType
+}
+
+export type ServingAuth = BaseServingAuth
+  & OneOf<{ tritonRestrictedKeyValue: ServingAuthTritonRestrictedKeyValue }>
+
+export type CreateInferenceServingRequest = {
+  workspace?: number
+  name?: string
+  cluster?: string
+  namespace?: string
+  models?: ServingConfig[]
+  serviceConfig?: ServiceConfig
+  replicas?: number
+  labels?: {[key: string]: string}
+  annotations?: {[key: string]: string}
+  podConfig?: BaizeCommonK8s.PodConfig
+  framework?: Framework
+  auth?: ServingAuth
+}
+
+export type UpdateInferenceServingRequest = {
+  workspace?: number
+  name?: string
+  cluster?: string
+  namespace?: string
+  models?: ServingConfig[]
+  replicas?: number
+  labels?: {[key: string]: string}
+  annotations?: {[key: string]: string}
+  podConfig?: BaizeCommonK8s.PodConfig
+  auth?: ServingAuth
+}
+
+export type SingleInferenceServingRequest = {
+  name?: string
+  cluster?: string
+  namespace?: string
+  workspace?: number
+}
+
+export type ListInferenceServingsRequest = {
+  cluster?: string
+  namespace?: string
+  workspace?: number
+  page?: BaizeCommonCommon.Pagination
+}
+
+export type ListInferenceServingsResponse = {
+  items?: InferenceServing[]
+  page?: BaizeCommonCommon.Pagination
+}
+
+export class InferenceServingManagement {
+  static CreateInferenceServing(req: CreateInferenceServingRequest, initReq?: fm.InitReq): Promise<InferenceServing> {
+    return fm.fetchReq<CreateInferenceServingRequest, InferenceServing>(`/apis/baize.io/v1alpha1/workspaces/${req["workspace"]}/clusters/${req["cluster"]}/namespaces/${req["namespace"]}/inference-serving`, {...initReq, method: "POST", body: JSON.stringify(req, fm.replacer)})
+  }
+  static GetInferenceServing(req: SingleInferenceServingRequest, initReq?: fm.InitReq): Promise<InferenceServing> {
+    return fm.fetchReq<SingleInferenceServingRequest, InferenceServing>(`/apis/baize.io/v1alpha1/workspaces/${req["workspace"]}/clusters/${req["cluster"]}/namespaces/${req["namespace"]}/inference-serving/${req["name"]}?${fm.renderURLSearchParams(req, ["workspace", "cluster", "namespace", "name"])}`, {...initReq, method: "GET"})
+  }
+  static ListInferenceServings(req: ListInferenceServingsRequest, initReq?: fm.InitReq): Promise<ListInferenceServingsResponse> {
+    return fm.fetchReq<ListInferenceServingsRequest, ListInferenceServingsResponse>(`/apis/baize.io/v1alpha1/workspaces/${req["workspace"]}/clusters/${req["cluster"]}/namespaces/${req["namespace"]}/inference-serving?${fm.renderURLSearchParams(req, ["workspace", "cluster", "namespace"])}`, {...initReq, method: "GET"})
+  }
+  static UpdateInferenceServing(req: UpdateInferenceServingRequest, initReq?: fm.InitReq): Promise<InferenceServing> {
+    return fm.fetchReq<UpdateInferenceServingRequest, InferenceServing>(`/apis/baize.io/v1alpha1/workspaces/${req["workspace"]}/clusters/${req["cluster"]}/namespaces/${req["namespace"]}/inference-serving/${req["name"]}`, {...initReq, method: "PUT", body: JSON.stringify(req, fm.replacer)})
+  }
+  static DeleteInferenceServing(req: SingleInferenceServingRequest, initReq?: fm.InitReq): Promise<InferenceServing> {
+    return fm.fetchReq<SingleInferenceServingRequest, InferenceServing>(`/apis/baize.io/v1alpha1/workspaces/${req["workspace"]}/clusters/${req["cluster"]}/namespaces/${req["namespace"]}/inference-serving/${req["name"]}`, {...initReq, method: "DELETE", body: JSON.stringify(req, fm.replacer)})
+  }
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@daocloud-proto/baize",
-  "version": "v0.103.3",
+  "version": "v0.104.1",
   "description": "",
   "author": "",
   "license": "ISC"