@daocloud-proto/infery 0.0.0-dev-62a40402 → 0.0.0-dev-1b4bc97d

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

@@ -76,13 +76,6 @@ export enum AvailabilityStatus {
   AVAILABILITY_STATUS_UNKNOWN = "AVAILABILITY_STATUS_UNKNOWN",
 }
 
-export enum ConditionStatus {
-  CONDITION_STATUS_UNSPECIFIED = "CONDITION_STATUS_UNSPECIFIED",
-  CONDITION_STATUS_TRUE = "CONDITION_STATUS_TRUE",
-  CONDITION_STATUS_FALSE = "CONDITION_STATUS_FALSE",
-  CONDITION_STATUS_UNKNOWN = "CONDITION_STATUS_UNKNOWN",
-}
-
 export enum ModelServingOriginType {
   MODEL_SERVING_ORIGIN_TYPE_UNSPECIFIED = "MODEL_SERVING_ORIGIN_TYPE_UNSPECIFIED",
   MODEL_SERVING_ORIGIN_TYPE_STRUCTURED = "MODEL_SERVING_ORIGIN_TYPE_STRUCTURED",
@@ -91,13 +84,6 @@ export enum ModelServingOriginType {
   MODEL_SERVING_ORIGIN_TYPE_EXTERNAL = "MODEL_SERVING_ORIGIN_TYPE_EXTERNAL",
 }
 
-export enum RoleCapability {
-  ROLE_CAPABILITY_UNSPECIFIED = "ROLE_CAPABILITY_UNSPECIFIED",
-  ROLE_CAPABILITY_ENTRY = "ROLE_CAPABILITY_ENTRY",
-  ROLE_CAPABILITY_ROUTING = "ROLE_CAPABILITY_ROUTING",
-  ROLE_CAPABILITY_KV_CACHE_OFFLOAD = "ROLE_CAPABILITY_KV_CACHE_OFFLOAD",
-}
-
 export enum ModelServingAction {
   MODEL_SERVING_ACTION_UNSPECIFIED = "MODEL_SERVING_ACTION_UNSPECIFIED",
   MODEL_SERVING_ACTION_START = "MODEL_SERVING_ACTION_START",
@@ -177,14 +163,6 @@ export type InferenceTemplateDeployment = {
   inferenceTemplateName?: string
 }
 
-export type StatusCondition = {
-  type?: string
-  status?: ConditionStatus
-  reason?: string
-  message?: string
-  lastTransitionTime?: GoogleProtobufTimestamp.Timestamp
-}
-
 export type ModelServingStatusSummary = {
   phase?: OperationPhase
   syncStatus?: SyncStatus
@@ -192,20 +170,11 @@ export type ModelServingStatusSummary = {
   availability?: AvailabilityStatus
 }
 
-export type RoleStatus = {
-  role?: RoleType
-  desiredReplicas?: number
-  readyReplicas?: number
-  restartingPods?: number
-}
-
 export type ModelServiceStatus = {
   phase?: OperationPhase
   syncStatus?: SyncStatus
   healthStatus?: HealthStatus
   availability?: AvailabilityStatus
-  roles?: RoleStatus[]
-  conditions?: StatusCondition[]
 }
 
 export type ModelServingOrigin = {
@@ -271,15 +240,6 @@ export type CreateModelServingResponse = {
 
 export type ListModelServingRequest = {
   workspace?: number
-  name?: string
-  phase?: OperationPhase
-  availability?: AvailabilityStatus
-  syncStatus?: SyncStatus
-  cluster?: string
-  namespace?: string
-  stack?: InferyCommonCommon.InferenceStack
-  disaggregationMode?: InferyCommonCommon.DisaggregationMode
-  gpuModel?: string
   page?: InferyCommonCommon.Pagination
 }
 
@@ -310,6 +270,20 @@ export type ListModelServingResponse = {
   page?: InferyCommonCommon.Pagination
 }
 
+
+type BaseRoleCapability = {
+}
+
+export type RoleCapability = BaseRoleCapability
+  & OneOf<{ kvCacheOffload: KVCacheOffloadConfig }>
+
+
+type BaseKVCacheOffloadConfig = {
+}
+
+export type KVCacheOffloadConfig = BaseKVCacheOffloadConfig
+  & OneOf<{ redhare: RedhareConfig }>
+
 export type GPUAllocation = {
   vendor?: string
   model?: string
@@ -319,17 +293,29 @@ export type GPUAllocation = {
   totalGpuCount?: number
 }
 
-export type ModelServingRoleOverview = {
+
+type BaseModelServingRoleOverview = {
   role?: RoleType
   componentName?: string
   desiredReplicas?: number
   readyReplicas?: number
   image?: string
+}
+
+export type ModelServingRoleOverview = BaseModelServingRoleOverview
+  & OneOf<{ worker: ModelServingWorkerOverview }>
+
+export type ModelServingWorkerOverview = {
   gpu?: GPUAllocation
   parallelism?: ParallelismConfig
   capabilities?: RoleCapability[]
 }
 
+export type RedhareConfig = {
+  redisUrl?: string
+  capacityBytes?: string
+}
+
 export type ModelServingAccess = {
   serviceAddress?: string
   accessPath?: string
@@ -340,6 +326,7 @@ export type GetModelServingResponse = {
   modelServing?: ModelServing
   topology?: ModelServingRoleOverview[]
   access?: ModelServingAccess
+  modelConfiguration?: ModelConfiguration
 }
 
 export type GetModelServingRequest = {
@@ -390,7 +377,6 @@ export type UpdateModelServingRequest = {
 export type DeleteModelServingRequest = {
   workspace?: number
   name?: string
-  resourceVersion?: string
 }
 
 export type ModelServingActionRequest = {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@daocloud-proto/infery",
-  "version": "v0.0.0-dev-62a40402",
+  "version": "v0.0.0-dev-1b4bc97d",
   "description": "Infery frontend protobuf SDK",
   "license": "ISC"
 }