@meetkai/mka1 0.49.17 → 0.49.18

package/src/models/operations/registerdeployment.ts

@@ -0,0 +1,69 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ */
+
+import * as z from "zod/v3";
+import { remap as remap$ } from "../../lib/primitives.js";
+import { safeParse } from "../../lib/schemas.js";
+import { Result as SafeParseResult } from "../../types/fp.js";
+import { SDKValidationError } from "../errors/sdkvalidationerror.js";
+
+export type RegisterDeploymentRequest = {
+  deploymentId: string;
+  /**
+   * Optional external end-user identifier forwarded by the API gateway.
+   */
+  xOnBehalfOf?: string | undefined;
+};
+
+/** @internal */
+export const RegisterDeploymentRequest$inboundSchema: z.ZodType<
+  RegisterDeploymentRequest,
+  z.ZodTypeDef,
+  unknown
+> = z.object({
+  deployment_id: z.string(),
+  "X-On-Behalf-Of": z.string().optional(),
+}).transform((v) => {
+  return remap$(v, {
+    "deployment_id": "deploymentId",
+    "X-On-Behalf-Of": "xOnBehalfOf",
+  });
+});
+/** @internal */
+export type RegisterDeploymentRequest$Outbound = {
+  deployment_id: string;
+  "X-On-Behalf-Of"?: string | undefined;
+};
+
+/** @internal */
+export const RegisterDeploymentRequest$outboundSchema: z.ZodType<
+  RegisterDeploymentRequest$Outbound,
+  z.ZodTypeDef,
+  RegisterDeploymentRequest
+> = z.object({
+  deploymentId: z.string(),
+  xOnBehalfOf: z.string().optional(),
+}).transform((v) => {
+  return remap$(v, {
+    deploymentId: "deployment_id",
+    xOnBehalfOf: "X-On-Behalf-Of",
+  });
+});
+
+export function registerDeploymentRequestToJSON(
+  registerDeploymentRequest: RegisterDeploymentRequest,
+): string {
+  return JSON.stringify(
+    RegisterDeploymentRequest$outboundSchema.parse(registerDeploymentRequest),
+  );
+}
+export function registerDeploymentRequestFromJSON(
+  jsonString: string,
+): SafeParseResult<RegisterDeploymentRequest, SDKValidationError> {
+  return safeParse(
+    jsonString,
+    (x) => RegisterDeploymentRequest$inboundSchema.parse(JSON.parse(x)),
+    `Failed to parse 'RegisterDeploymentRequest' from JSON`,
+  );
+}

package/src/react-query/index.ts

@@ -294,12 +294,14 @@ export * from "./sandboxUploadFile.js";
 export * from "./servingAcceleratorsList.js";
 export * from "./servingDeploymentsCreate.js";
 export * from "./servingDeploymentsDelete.js";
+export * from "./servingDeploymentsDeregister.js";
 export * from "./servingDeploymentsGet.js";
 export * from "./servingDeploymentsGetLogs.js";
 export * from "./servingDeploymentsGetMetrics.js";
 export * from "./servingDeploymentsList.js";
 export * from "./servingDeploymentsListReplicas.js";
 export * from "./servingDeploymentsListRevisions.js";
+export * from "./servingDeploymentsRegister.js";
 export * from "./servingDeploymentsRollback.js";
 export * from "./servingDeploymentsUpdate.js";
 export * from "./servingDeploymentsUpdateScaling.js";

package/src/react-query/servingDeploymentsDeregister.ts

@@ -0,0 +1,111 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ */
+
+import {
+  MutationKey,
+  useMutation,
+  UseMutationResult,
+} from "@tanstack/react-query";
+import { SDKCore } from "../core.js";
+import { servingDeploymentsDeregister } from "../funcs/servingDeploymentsDeregister.js";
+import { combineSignals } from "../lib/primitives.js";
+import { RequestOptions } from "../lib/sdks.js";
+import {
+  ConnectionError,
+  InvalidRequestError,
+  RequestAbortedError,
+  RequestTimeoutError,
+  UnexpectedClientError,
+} from "../models/errors/httpclienterrors.js";
+import * as errors from "../models/errors/index.js";
+import { ResponseValidationError } from "../models/errors/responsevalidationerror.js";
+import { SDKError } from "../models/errors/sdkerror.js";
+import { SDKValidationError } from "../models/errors/sdkvalidationerror.js";
+import * as operations from "../models/operations/index.js";
+import { unwrapAsync } from "../types/fp.js";
+import { useSDKContext } from "./_context.js";
+import { MutationHookOptions } from "./_types.js";
+
+export type ServingDeploymentsDeregisterMutationVariables = {
+  request: operations.DeregisterDeploymentRequest;
+  options?: RequestOptions;
+};
+
+export type ServingDeploymentsDeregisterMutationData = void;
+
+export type ServingDeploymentsDeregisterMutationError =
+  | errors.ServingError
+  | errors.ServingHTTPValidationError
+  | SDKError
+  | ResponseValidationError
+  | ConnectionError
+  | RequestAbortedError
+  | RequestTimeoutError
+  | InvalidRequestError
+  | UnexpectedClientError
+  | SDKValidationError;
+
+/**
+ * Deregister the model from the LLM gateway
+ *
+ * @remarks
+ * Withdraw the deployment's model from the LLM gateway registry: inference stops resolving, compute keeps running (re-register any time).
+ */
+export function useServingDeploymentsDeregisterMutation(
+  options?: MutationHookOptions<
+    ServingDeploymentsDeregisterMutationData,
+    ServingDeploymentsDeregisterMutationError,
+    ServingDeploymentsDeregisterMutationVariables
+  >,
+): UseMutationResult<
+  ServingDeploymentsDeregisterMutationData,
+  ServingDeploymentsDeregisterMutationError,
+  ServingDeploymentsDeregisterMutationVariables
+> {
+  const client = useSDKContext();
+  return useMutation({
+    ...buildServingDeploymentsDeregisterMutation(client, options),
+    ...options,
+  });
+}
+
+export function mutationKeyServingDeploymentsDeregister(): MutationKey {
+  return ["@meetkai/mka1", "deployments", "deregister"];
+}
+
+export function buildServingDeploymentsDeregisterMutation(
+  client$: SDKCore,
+  hookOptions?: RequestOptions,
+): {
+  mutationKey: MutationKey;
+  mutationFn: (
+    variables: ServingDeploymentsDeregisterMutationVariables,
+  ) => Promise<ServingDeploymentsDeregisterMutationData>;
+} {
+  return {
+    mutationKey: mutationKeyServingDeploymentsDeregister(),
+    mutationFn: function servingDeploymentsDeregisterMutationFn({
+      request,
+      options,
+    }): Promise<ServingDeploymentsDeregisterMutationData> {
+      const mergedOptions = {
+        ...hookOptions,
+        ...options,
+        fetchOptions: {
+          ...hookOptions?.fetchOptions,
+          ...options?.fetchOptions,
+          signal: combineSignals(
+            hookOptions?.fetchOptions?.signal,
+            options?.fetchOptions?.signal,
+          ),
+        },
+      };
+      return unwrapAsync(servingDeploymentsDeregister(
+        client$,
+        request,
+        mergedOptions,
+      ));
+    },
+  };
+}

package/src/react-query/servingDeploymentsRegister.ts

@@ -0,0 +1,112 @@
+/*
+ * Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT.
+ */
+
+import {
+  MutationKey,
+  useMutation,
+  UseMutationResult,
+} from "@tanstack/react-query";
+import { SDKCore } from "../core.js";
+import { servingDeploymentsRegister } from "../funcs/servingDeploymentsRegister.js";
+import { combineSignals } from "../lib/primitives.js";
+import { RequestOptions } from "../lib/sdks.js";
+import * as components from "../models/components/index.js";
+import {
+  ConnectionError,
+  InvalidRequestError,
+  RequestAbortedError,
+  RequestTimeoutError,
+  UnexpectedClientError,
+} from "../models/errors/httpclienterrors.js";
+import * as errors from "../models/errors/index.js";
+import { ResponseValidationError } from "../models/errors/responsevalidationerror.js";
+import { SDKError } from "../models/errors/sdkerror.js";
+import { SDKValidationError } from "../models/errors/sdkvalidationerror.js";
+import * as operations from "../models/operations/index.js";
+import { unwrapAsync } from "../types/fp.js";
+import { useSDKContext } from "./_context.js";
+import { MutationHookOptions } from "./_types.js";
+
+export type ServingDeploymentsRegisterMutationVariables = {
+  request: operations.RegisterDeploymentRequest;
+  options?: RequestOptions;
+};
+
+export type ServingDeploymentsRegisterMutationData = components.Deployment;
+
+export type ServingDeploymentsRegisterMutationError =
+  | errors.ServingError
+  | errors.ServingHTTPValidationError
+  | SDKError
+  | ResponseValidationError
+  | ConnectionError
+  | RequestAbortedError
+  | RequestTimeoutError
+  | InvalidRequestError
+  | UnexpectedClientError
+  | SDKValidationError;
+
+/**
+ * Register the model with the LLM gateway
+ *
+ * @remarks
+ * Make the deployment's model available for inference: registers it in the LLM gateway's model registry under the deployment's name. Deployment must be ready. Registration is deliberately decoupled from deployment so pricing, budgets, and rate limits can be configured before the model becomes callable.
+ */
+export function useServingDeploymentsRegisterMutation(
+  options?: MutationHookOptions<
+    ServingDeploymentsRegisterMutationData,
+    ServingDeploymentsRegisterMutationError,
+    ServingDeploymentsRegisterMutationVariables
+  >,
+): UseMutationResult<
+  ServingDeploymentsRegisterMutationData,
+  ServingDeploymentsRegisterMutationError,
+  ServingDeploymentsRegisterMutationVariables
+> {
+  const client = useSDKContext();
+  return useMutation({
+    ...buildServingDeploymentsRegisterMutation(client, options),
+    ...options,
+  });
+}
+
+export function mutationKeyServingDeploymentsRegister(): MutationKey {
+  return ["@meetkai/mka1", "deployments", "register"];
+}
+
+export function buildServingDeploymentsRegisterMutation(
+  client$: SDKCore,
+  hookOptions?: RequestOptions,
+): {
+  mutationKey: MutationKey;
+  mutationFn: (
+    variables: ServingDeploymentsRegisterMutationVariables,
+  ) => Promise<ServingDeploymentsRegisterMutationData>;
+} {
+  return {
+    mutationKey: mutationKeyServingDeploymentsRegister(),
+    mutationFn: function servingDeploymentsRegisterMutationFn({
+      request,
+      options,
+    }): Promise<ServingDeploymentsRegisterMutationData> {
+      const mergedOptions = {
+        ...hookOptions,
+        ...options,
+        fetchOptions: {
+          ...hookOptions?.fetchOptions,
+          ...options?.fetchOptions,
+          signal: combineSignals(
+            hookOptions?.fetchOptions?.signal,
+            options?.fetchOptions?.signal,
+          ),
+        },
+      };
+      return unwrapAsync(servingDeploymentsRegister(
+        client$,
+        request,
+        mergedOptions,
+      ));
+    },
+  };
+}

package/src/react-query/servingFinetuneJobsGetEvents.core.ts

@@ -43,6 +43,7 @@ export function buildServingFinetuneJobsGetEventsQuery(
 } {
   return {
     queryKey: queryKeyServingFinetuneJobsGetEvents(request.jobId, {
+      includeLineage: request.includeLineage,
       xOnBehalfOf: request.xOnBehalfOf,
     }),
     queryFn: async function servingFinetuneJobsGetEventsQueryFn(
@@ -70,7 +71,10 @@ export function buildServingFinetuneJobsGetEventsQuery(
 
 export function queryKeyServingFinetuneJobsGetEvents(
   jobId: string,
-  parameters: { xOnBehalfOf?: string | undefined },
+  parameters: {
+    includeLineage?: boolean | undefined;
+    xOnBehalfOf?: string | undefined;
+  },
 ): QueryKey {
   return ["@meetkai/mka1", "finetuneJobs", "getEvents", jobId, parameters];
 }

package/src/react-query/servingFinetuneJobsGetEvents.ts

@@ -111,7 +111,10 @@ export function setServingFinetuneJobsGetEventsData(
   client: QueryClient,
   queryKeyBase: [
     jobId: string,
-    parameters: { xOnBehalfOf?: string | undefined },
+    parameters: {
+      includeLineage?: boolean | undefined;
+      xOnBehalfOf?: string | undefined;
+    },
   ],
   data: ServingFinetuneJobsGetEventsQueryData,
 ): ServingFinetuneJobsGetEventsQueryData | undefined {
@@ -123,7 +126,13 @@ export function setServingFinetuneJobsGetEventsData(
 export function invalidateServingFinetuneJobsGetEvents(
   client: QueryClient,
   queryKeyBase: TupleToPrefixes<
-    [jobId: string, parameters: { xOnBehalfOf?: string | undefined }]
+    [
+      jobId: string,
+      parameters: {
+        includeLineage?: boolean | undefined;
+        xOnBehalfOf?: string | undefined;
+      },
+    ]
   >,
   filters?: Omit<InvalidateQueryFilters, "queryKey" | "predicate" | "exact">,
 ): Promise<void> {

package/src/sdk/deployments.ts

@@ -4,12 +4,14 @@
 
 import { servingDeploymentsCreate } from "../funcs/servingDeploymentsCreate.js";
 import { servingDeploymentsDelete } from "../funcs/servingDeploymentsDelete.js";
+import { servingDeploymentsDeregister } from "../funcs/servingDeploymentsDeregister.js";
 import { servingDeploymentsGet } from "../funcs/servingDeploymentsGet.js";
 import { servingDeploymentsGetLogs } from "../funcs/servingDeploymentsGetLogs.js";
 import { servingDeploymentsGetMetrics } from "../funcs/servingDeploymentsGetMetrics.js";
 import { servingDeploymentsList } from "../funcs/servingDeploymentsList.js";
 import { servingDeploymentsListReplicas } from "../funcs/servingDeploymentsListReplicas.js";
 import { servingDeploymentsListRevisions } from "../funcs/servingDeploymentsListRevisions.js";
+import { servingDeploymentsRegister } from "../funcs/servingDeploymentsRegister.js";
 import { servingDeploymentsRollback } from "../funcs/servingDeploymentsRollback.js";
 import { servingDeploymentsUpdate } from "../funcs/servingDeploymentsUpdate.js";
 import { servingDeploymentsUpdateScaling } from "../funcs/servingDeploymentsUpdateScaling.js";
@@ -19,6 +21,23 @@ import * as operations from "../models/operations/index.js";
 import { unwrapAsync } from "../types/fp.js";
 
 export class Deployments extends ClientSDK {
+  /**
+   * List deployments
+   *
+   * @remarks
+   * List your inference deployments.
+   */
+  async list(
+    request: operations.ListDeploymentsRequest,
+    options?: RequestOptions,
+  ): Promise<components.PageDeployment> {
+    return unwrapAsync(servingDeploymentsList(
+      this,
+      request,
+      options,
+    ));
+  }
+
   /**
    * Create an inference deployment
    *
@@ -37,16 +56,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * List deployments
+   * Tear down a deployment
    *
    * @remarks
-   * List your inference deployments.
+   * Tear down a deployment and release its resources.
    */
-  async list(
-    request: operations.ListDeploymentsRequest,
+  async delete(
+    request: operations.DeleteDeploymentRequest,
     options?: RequestOptions,
-  ): Promise<components.PageDeployment> {
-    return unwrapAsync(servingDeploymentsList(
+  ): Promise<void> {
+    return unwrapAsync(servingDeploymentsDelete(
       this,
       request,
       options,
@@ -88,16 +107,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * Tear down a deployment
+   * Fetch deployment logs
    *
    * @remarks
-   * Tear down a deployment and release its resources.
+   * Fetch recent log lines from a deployment's replicas.
    */
-  async delete(
-    request: operations.DeleteDeploymentRequest,
+  async getLogs(
+    request: operations.GetDeploymentLogsRequest,
     options?: RequestOptions,
-  ): Promise<void> {
-    return unwrapAsync(servingDeploymentsDelete(
+  ): Promise<Array<components.LogLine>> {
+    return unwrapAsync(servingDeploymentsGetLogs(
       this,
       request,
       options,
@@ -105,16 +124,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * Update autoscaling live (no redeploy)
+   * Live serving metrics
    *
    * @remarks
-   * Change a deployment's autoscaling limits in place, without redeploying.
+   * Return live serving metrics (throughput, queue depth, accelerator usage).
    */
-  async updateScaling(
-    request: operations.UpdateScalingRequest,
+  async getMetrics(
+    request: operations.GetDeploymentMetricsRequest,
     options?: RequestOptions,
-  ): Promise<components.Deployment> {
-    return unwrapAsync(servingDeploymentsUpdateScaling(
+  ): Promise<components.Metrics> {
+    return unwrapAsync(servingDeploymentsGetMetrics(
       this,
       request,
       options,
@@ -122,16 +141,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * List config revisions
+   * Deregister the model from the LLM gateway
    *
    * @remarks
-   * List a deployment's configuration revisions, most recent first.
+   * Withdraw the deployment's model from the LLM gateway registry: inference stops resolving, compute keeps running (re-register any time).
    */
-  async listRevisions(
-    request: operations.ListRevisionsRequest,
+  async deregister(
+    request: operations.DeregisterDeploymentRequest,
     options?: RequestOptions,
-  ): Promise<components.PageRevision> {
-    return unwrapAsync(servingDeploymentsListRevisions(
+  ): Promise<void> {
+    return unwrapAsync(servingDeploymentsDeregister(
       this,
       request,
       options,
@@ -139,16 +158,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * Roll back to a previous revision
+   * Register the model with the LLM gateway
    *
    * @remarks
-   * Roll a deployment back to a previous configuration revision.
+   * Make the deployment's model available for inference: registers it in the LLM gateway's model registry under the deployment's name. Deployment must be ready. Registration is deliberately decoupled from deployment so pricing, budgets, and rate limits can be configured before the model becomes callable.
    */
-  async rollback(
-    request: operations.RollbackDeploymentRequest,
+  async register(
+    request: operations.RegisterDeploymentRequest,
     options?: RequestOptions,
   ): Promise<components.Deployment> {
-    return unwrapAsync(servingDeploymentsRollback(
+    return unwrapAsync(servingDeploymentsRegister(
       this,
       request,
       options,
@@ -156,16 +175,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * Fetch deployment logs
+   * List running replicas
    *
    * @remarks
-   * Fetch recent log lines from a deployment's replicas.
+   * List the replicas currently backing a deployment.
    */
-  async getLogs(
-    request: operations.GetDeploymentLogsRequest,
+  async listReplicas(
+    request: operations.ListReplicasRequest,
     options?: RequestOptions,
-  ): Promise<Array<components.LogLine>> {
-    return unwrapAsync(servingDeploymentsGetLogs(
+  ): Promise<components.PageReplica> {
+    return unwrapAsync(servingDeploymentsListReplicas(
       this,
       request,
       options,
@@ -173,16 +192,16 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * Live serving metrics
+   * List config revisions
    *
    * @remarks
-   * Return live serving metrics (throughput, queue depth, accelerator usage).
+   * List a deployment's configuration revisions, most recent first.
    */
-  async getMetrics(
-    request: operations.GetDeploymentMetricsRequest,
+  async listRevisions(
+    request: operations.ListRevisionsRequest,
     options?: RequestOptions,
-  ): Promise<components.Metrics> {
-    return unwrapAsync(servingDeploymentsGetMetrics(
+  ): Promise<components.PageRevision> {
+    return unwrapAsync(servingDeploymentsListRevisions(
       this,
       request,
       options,
@@ -190,16 +209,33 @@ export class Deployments extends ClientSDK {
   }
 
   /**
-   * List running replicas
+   * Roll back to a previous revision
    *
    * @remarks
-   * List the replicas currently backing a deployment.
+   * Roll a deployment back to a previous configuration revision.
    */
-  async listReplicas(
-    request: operations.ListReplicasRequest,
+  async rollback(
+    request: operations.RollbackDeploymentRequest,
     options?: RequestOptions,
-  ): Promise<components.PageReplica> {
-    return unwrapAsync(servingDeploymentsListReplicas(
+  ): Promise<components.Deployment> {
+    return unwrapAsync(servingDeploymentsRollback(
+      this,
+      request,
+      options,
+    ));
+  }
+
+  /**
+   * Update autoscaling live (no redeploy)
+   *
+   * @remarks
+   * Change a deployment's autoscaling limits in place, without redeploying.
+   */
+  async updateScaling(
+    request: operations.UpdateScalingRequest,
+    options?: RequestOptions,
+  ): Promise<components.Deployment> {
+    return unwrapAsync(servingDeploymentsUpdateScaling(
       this,
       request,
       options,