llama-stack-client 0.4.4 → 0.4.5

package/dist/src/resources/vector-stores/files.ts

@@ -13,7 +13,7 @@ import { OpenAICursorPage, type OpenAICursorPageParams } from '../../pagination'
 
 export class Files extends APIResource {
   /**
-   * Attach a file to a vector store.
+   * Attach a file to a vector store (OpenAI-compatible).
    */
   create(
     vectorStoreId: string,
@@ -24,7 +24,7 @@ export class Files extends APIResource {
   }
 
   /**
-   * Retrieves a vector store file.
+   * Retrieve a vector store file (OpenAI-compatible).
    */
   retrieve(
     vectorStoreId: string,
@@ -35,7 +35,7 @@ export class Files extends APIResource {
   }
 
   /**
-   * Updates a vector store file.
+   * Update a vector store file (OpenAI-compatible).
    */
   update(
     vectorStoreId: string,
@@ -47,7 +47,7 @@ export class Files extends APIResource {
   }
 
   /**
-   * List files in a vector store.
+   * List files in a vector store (OpenAI-compatible).
    */
   list(
     vectorStoreId: string,
@@ -74,7 +74,7 @@ export class Files extends APIResource {
   }
 
   /**
-   * Delete a vector store file.
+   * Delete a vector store file (OpenAI-compatible).
    */
   delete(
     vectorStoreId: string,
@@ -85,7 +85,7 @@ export class Files extends APIResource {
   }
 
   /**
-   * Retrieves the contents of a vector store file.
+   * Retrieve vector store file contents (OpenAI-compatible).
    */
   content(
     vectorStoreId: string,
@@ -253,10 +253,6 @@ export namespace FileContentResponse {
      * the context during inference.
      */
     export interface ChunkMetadata {
-      chunk_embedding_dimension?: number | null;
-
-      chunk_embedding_model?: string | null;
-
       chunk_id?: string | null;
 
       chunk_tokenizer?: string | null;
@@ -279,12 +275,18 @@ export namespace FileContentResponse {
 }
 
 export interface FileCreateParams {
+  /**
+   * The ID of the file to attach.
+   */
   file_id: string;
 
+  /**
+   * Attributes to associate with the file.
+   */
   attributes?: { [key: string]: unknown } | null;
 
   /**
-   * Automatic chunking strategy for vector store files.
+   * Strategy for chunking the file content.
    */
   chunking_strategy?:
     | FileCreateParams.VectorStoreChunkingStrategyAuto
@@ -325,20 +327,38 @@ export namespace FileCreateParams {
 }
 
 export interface FileUpdateParams {
+  /**
+   * The new attributes for the file.
+   */
   attributes: { [key: string]: unknown };
 }
 
 export interface FileListParams extends OpenAICursorPageParams {
+  /**
+   * Pagination cursor (before).
+   */
   before?: string | null;
 
+  /**
+   * Filter by file status.
+   */
   filter?: 'completed' | 'in_progress' | 'cancelled' | 'failed' | null;
 
+  /**
+   * Sort order by created_at: asc or desc.
+   */
   order?: string | null;
 }
 
 export interface FileContentParams {
+  /**
+   * Include embedding vectors.
+   */
   include_embeddings?: boolean | null;
 
+  /**
+   * Include chunk metadata.
+   */
   include_metadata?: boolean | null;
 }
 

package/dist/src/resources/vector-stores/vector-stores.ts

@@ -36,23 +36,21 @@ export class VectorStores extends APIResource {
   fileBatches: FileBatchesAPI.FileBatches = new FileBatchesAPI.FileBatches(this._client);
 
   /**
-   * Creates a vector store.
-   *
-   * Generate an OpenAI-compatible vector store with the given parameters.
+   * Create a vector store (OpenAI-compatible).
    */
   create(body: VectorStoreCreateParams, options?: Core.RequestOptions): Core.APIPromise<VectorStore> {
     return this._client.post('/v1/vector_stores', { body, ...options });
   }
 
   /**
-   * Retrieves a vector store.
+   * Retrieve a vector store (OpenAI-compatible).
    */
   retrieve(vectorStoreId: string, options?: Core.RequestOptions): Core.APIPromise<VectorStore> {
     return this._client.get(`/v1/vector_stores/${vectorStoreId}`, options);
   }
 
   /**
-   * Updates a vector store.
+   * Update a vector store (OpenAI-compatible).
    */
   update(
     vectorStoreId: string,
@@ -63,7 +61,7 @@ export class VectorStores extends APIResource {
   }
 
   /**
-   * Returns a list of vector stores.
+   * List vector stores (OpenAI-compatible).
    */
   list(
     query?: VectorStoreListParams,
@@ -81,17 +79,14 @@ export class VectorStores extends APIResource {
   }
 
   /**
-   * Delete a vector store.
+   * Delete a vector store (OpenAI-compatible).
    */
   delete(vectorStoreId: string, options?: Core.RequestOptions): Core.APIPromise<VectorStoreDeleteResponse> {
     return this._client.delete(`/v1/vector_stores/${vectorStoreId}`, options);
   }
 
   /**
-   * Search for chunks in a vector store.
-   *
-   * Searches a vector store for relevant chunks based on a query and optional file
-   * attribute filters.
+   * Search a vector store (OpenAI-compatible).
    */
   search(
     vectorStoreId: string,
@@ -242,10 +237,6 @@ export namespace VectorStoreSearchResponse {
        * the context during inference.
        */
       export interface ChunkMetadata {
-        chunk_embedding_dimension?: number | null;
-
-        chunk_embedding_model?: string | null;
-
         chunk_id?: string | null;
 
         chunk_tokenizer?: string | null;
@@ -321,44 +312,127 @@ export namespace VectorStoreCreateParams {
 }
 
 export interface VectorStoreUpdateParams {
+  /**
+   * Expiration policy for the vector store.
+   */
   expires_after?: { [key: string]: unknown } | null;
 
+  /**
+   * Metadata to associate with the vector store.
+   */
   metadata?: { [key: string]: unknown } | null;
 
+  /**
+   * The new name for the vector store.
+   */
   name?: string | null;
 }
 
 export interface VectorStoreListParams extends OpenAICursorPageParams {
+  /**
+   * Pagination cursor (before).
+   */
   before?: string | null;
 
+  /**
+   * Sort order by created_at: asc or desc.
+   */
   order?: string | null;
 }
 
 export interface VectorStoreSearchParams {
+  /**
+   * The search query string or list of query strings.
+   */
   query: string | Array<string>;
 
+  /**
+   * Filters to apply to the search.
+   */
   filters?: { [key: string]: unknown } | null;
 
+  /**
+   * Maximum number of results to return.
+   */
   max_num_results?: number | null;
 
   /**
    * Options for ranking and filtering search results.
+   *
+   * This class configures how search results are ranked and filtered. You can use
+   * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
+   * VectorStoresConfig are used when parameters are not provided.
+   *
+   * Examples: # Weighted ranker with custom alpha
+   * SearchRankingOptions(ranker="weighted", alpha=0.7)
+   *
+   *     # RRF ranker with custom impact factor
+   *     SearchRankingOptions(ranker="rrf", impact_factor=50.0)
+   *
+   *     # Use config defaults (just specify ranker type)
+   *     SearchRankingOptions(ranker="weighted")  # Uses alpha from VectorStoresConfig
+   *
+   *     # Score threshold filtering
+   *     SearchRankingOptions(ranker="weighted", score_threshold=0.5)
    */
   ranking_options?: VectorStoreSearchParams.RankingOptions | null;
 
+  /**
+   * Whether to rewrite the query for better results.
+   */
   rewrite_query?: boolean | null;
 
+  /**
+   * The search mode to use (e.g., 'vector', 'keyword').
+   */
   search_mode?: string | null;
 }
 
 export namespace VectorStoreSearchParams {
   /**
    * Options for ranking and filtering search results.
+   *
+   * This class configures how search results are ranked and filtered. You can use
+   * algorithm-based rerankers (weighted, RRF) or neural rerankers. Defaults from
+   * VectorStoresConfig are used when parameters are not provided.
+   *
+   * Examples: # Weighted ranker with custom alpha
+   * SearchRankingOptions(ranker="weighted", alpha=0.7)
+   *
+   *     # RRF ranker with custom impact factor
+   *     SearchRankingOptions(ranker="rrf", impact_factor=50.0)
+   *
+   *     # Use config defaults (just specify ranker type)
+   *     SearchRankingOptions(ranker="weighted")  # Uses alpha from VectorStoresConfig
+   *
+   *     # Score threshold filtering
+   *     SearchRankingOptions(ranker="weighted", score_threshold=0.5)
    */
   export interface RankingOptions {
+    /**
+     * Weight factor for weighted ranker
+     */
+    alpha?: number | null;
+
+    /**
+     * Impact factor for RRF algorithm
+     */
+    impact_factor?: number | null;
+
+    /**
+     * Model identifier for neural reranker
+     */
+    model?: string | null;
+
     ranker?: string | null;
 
     score_threshold?: number | null;
+
+    /**
+     * Weights for combining vector, keyword, and neural scores. Keys: 'vector',
+     * 'keyword', 'neural'
+     */
+    weights?: { [key: string]: number } | null;
   }
 }
 

package/dist/src/version.ts

@@ -4,4 +4,4 @@
 // This source code is licensed under the terms described in the LICENSE file in
 // the root directory of this source tree.
 
-export const VERSION = '0.4.4'; // x-release-please-version
+export const VERSION = '0.4.5'; // x-release-please-version

package/dist/version.d.ts

@@ -1,2 +1,2 @@
-export declare const VERSION = "0.4.4";
+export declare const VERSION = "0.4.5";
 //# sourceMappingURL=version.d.ts.map

package/dist/version.js

@@ -6,5 +6,5 @@
 // the root directory of this source tree.
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.VERSION = void 0;
-exports.VERSION = '0.4.4'; // x-release-please-version
+exports.VERSION = '0.4.5'; // x-release-please-version
 //# sourceMappingURL=version.js.map

package/dist/version.mjs

@@ -3,5 +3,5 @@
 //
 // This source code is licensed under the terms described in the LICENSE file in
 // the root directory of this source tree.
-export const VERSION = '0.4.4'; // x-release-please-version
+export const VERSION = '0.4.5'; // x-release-please-version
 //# sourceMappingURL=version.mjs.map

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llama-stack-client",
-  "version": "0.4.4",
+  "version": "0.4.5",
   "description": "The official TypeScript library for the Llama Stack Client API",
   "author": "Llama Stack Client <llamastack@meta.com>",
   "types": "dist/index.d.ts",

package/src/core.ts

@@ -90,6 +90,12 @@ async function defaultParseResponse<T>(props: APIResponseProps): Promise<T> {
   const mediaType = contentType?.split(';')[0]?.trim();
   const isJSON = mediaType?.includes('application/json') || mediaType?.endsWith('+json');
   if (isJSON) {
+    const contentLength = response.headers.get('content-length');
+    if (contentLength === '0') {
+      // if there is no content we can't do anything
+      return undefined as T;
+    }
+
     const json = await response.json();
 
     debug('response', response.status, response.url, response.headers, json);

package/src/index.ts

@@ -48,10 +48,10 @@ import {
   FilesOpenAICursorPage,
   ListFilesResponse,
 } from './resources/files';
-import { HealthInfo, Inspect, ProviderInfo, RouteInfo, VersionInfo } from './resources/inspect';
+import { Inspect } from './resources/inspect';
 import { CreateResponse, ModerationCreateParams, Moderations } from './resources/moderations';
-import { ListProvidersResponse, ProviderListResponse, Providers } from './resources/providers';
-import { ListRoutesResponse, RouteListParams, RouteListResponse, Routes } from './resources/routes';
+import { ProviderListResponse, Providers } from './resources/providers';
+import { RouteListParams, RouteListResponse, Routes } from './resources/routes';
 import { RunShieldResponse, Safety, SafetyRunShieldParams } from './resources/safety';
 import {
   Scoring,
@@ -431,13 +431,7 @@ export declare namespace LlamaStackClient {
     type ConversationUpdateParams as ConversationUpdateParams,
   };
 
-  export {
-    Inspect as Inspect,
-    type HealthInfo as HealthInfo,
-    type ProviderInfo as ProviderInfo,
-    type RouteInfo as RouteInfo,
-    type VersionInfo as VersionInfo,
-  };
+  export { Inspect as Inspect };
 
   export {
     Embeddings as Embeddings,
@@ -485,15 +479,10 @@ export declare namespace LlamaStackClient {
     type ModelRegisterParams as ModelRegisterParams,
   };
 
-  export {
-    Providers as Providers,
-    type ListProvidersResponse as ListProvidersResponse,
-    type ProviderListResponse as ProviderListResponse,
-  };
+  export { Providers as Providers, type ProviderListResponse as ProviderListResponse };
 
   export {
     Routes as Routes,
-    type ListRoutesResponse as ListRoutesResponse,
     type RouteListResponse as RouteListResponse,
     type RouteListParams as RouteListParams,
   };
@@ -561,13 +550,19 @@ export declare namespace LlamaStackClient {
 
   export { Beta as Beta };
 
+  export type HealthInfo = API.HealthInfo;
   export type InterleavedContent = API.InterleavedContent;
   export type InterleavedContentItem = API.InterleavedContentItem;
+  export type ListProvidersResponse = API.ListProvidersResponse;
+  export type ListRoutesResponse = API.ListRoutesResponse;
   export type ParamType = API.ParamType;
+  export type ProviderInfo = API.ProviderInfo;
+  export type RouteInfo = API.RouteInfo;
   export type SafetyViolation = API.SafetyViolation;
   export type SamplingParams = API.SamplingParams;
   export type ScoringResult = API.ScoringResult;
   export type SystemMessage = API.SystemMessage;
+  export type VersionInfo = API.VersionInfo;
 }
 
 export { getResponseOutputText } from './lib/response-helpers';

package/src/resources/alpha/admin.ts

@@ -0,0 +1,84 @@
+// Copyright (c) Meta Platforms, Inc. and affiliates.
+// All rights reserved.
+//
+// This source code is licensed under the terms described in the LICENSE file in
+// the root directory of this source tree.
+//
+// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+import { APIResource } from '../../resource';
+import { isRequestOptions } from '../../core';
+import * as Core from '../../core';
+import * as ProvidersAPI from '../providers';
+import * as RoutesAPI from '../routes';
+import * as Shared from '../shared';
+
+export class Admin extends APIResource {
+  /**
+   * Get the current health status of the service.
+   */
+  health(options?: Core.RequestOptions): Core.APIPromise<Shared.HealthInfo> {
+    return this._client.get('/v1alpha/admin/health', options);
+  }
+
+  /**
+   * Get detailed information about a specific provider.
+   */
+  inspectProvider(providerId: string, options?: Core.RequestOptions): Core.APIPromise<Shared.ProviderInfo> {
+    return this._client.get(`/v1alpha/admin/providers/${providerId}`, options);
+  }
+
+  /**
+   * List all available providers with their configuration and health status.
+   */
+  listProviders(options?: Core.RequestOptions): Core.APIPromise<ProvidersAPI.ProviderListResponse> {
+    return (
+      this._client.get('/v1alpha/admin/providers', options) as Core.APIPromise<{
+        data: ProvidersAPI.ProviderListResponse;
+      }>
+    )._thenUnwrap((obj) => obj.data);
+  }
+
+  /**
+   * List all available API routes with their methods and implementing providers.
+   */
+  listRoutes(
+    query?: AdminListRoutesParams,
+    options?: Core.RequestOptions,
+  ): Core.APIPromise<RoutesAPI.RouteListResponse>;
+  listRoutes(options?: Core.RequestOptions): Core.APIPromise<RoutesAPI.RouteListResponse>;
+  listRoutes(
+    query: AdminListRoutesParams | Core.RequestOptions = {},
+    options?: Core.RequestOptions,
+  ): Core.APIPromise<RoutesAPI.RouteListResponse> {
+    if (isRequestOptions(query)) {
+      return this.listRoutes({}, query);
+    }
+    return (
+      this._client.get('/v1alpha/admin/inspect/routes', { query, ...options }) as Core.APIPromise<{
+        data: RoutesAPI.RouteListResponse;
+      }>
+    )._thenUnwrap((obj) => obj.data);
+  }
+
+  /**
+   * Get the version of the service.
+   */
+  version(options?: Core.RequestOptions): Core.APIPromise<Shared.VersionInfo> {
+    return this._client.get('/v1alpha/admin/version', options);
+  }
+}
+
+export interface AdminListRoutesParams {
+  /**
+   * Filter to control which routes are returned. Can be an API level ('v1',
+   * 'v1alpha', 'v1beta') to show non-deprecated routes at that level, or
+   * 'deprecated' to show deprecated routes across all levels. If not specified,
+   * returns all non-deprecated routes.
+   */
+  api_filter?: 'v1' | 'v1alpha' | 'v1beta' | 'deprecated' | null;
+}
+
+export declare namespace Admin {
+  export { type AdminListRoutesParams as AdminListRoutesParams };
+}

package/src/resources/alpha/alpha.ts

@@ -7,6 +7,8 @@
 // File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 import { APIResource } from '../../resource';
+import * as AdminAPI from './admin';
+import { Admin, AdminListRoutesParams } from './admin';
 import * as BenchmarksAPI from './benchmarks';
 import {
   Benchmark,
@@ -31,6 +33,7 @@ import {
 import * as PostTrainingAPI from './post-training/post-training';
 import {
   AlgorithmConfig,
+  ListPostTrainingJobsResponse,
   PostTraining,
   PostTrainingJob,
   PostTrainingPreferenceOptimizeParams,
@@ -38,27 +41,24 @@ import {
 } from './post-training/post-training';
 
 export class Alpha extends APIResource {
-  inference: InferenceAPI.Inference = new InferenceAPI.Inference(this._client);
   postTraining: PostTrainingAPI.PostTraining = new PostTrainingAPI.PostTraining(this._client);
   benchmarks: BenchmarksAPI.Benchmarks = new BenchmarksAPI.Benchmarks(this._client);
   eval: EvalAPI.Eval = new EvalAPI.Eval(this._client);
+  admin: AdminAPI.Admin = new AdminAPI.Admin(this._client);
+  inference: InferenceAPI.Inference = new InferenceAPI.Inference(this._client);
 }
 
-Alpha.Inference = Inference;
 Alpha.PostTraining = PostTraining;
 Alpha.Benchmarks = Benchmarks;
 Alpha.Eval = Eval;
+Alpha.Admin = Admin;
+Alpha.Inference = Inference;
 
 export declare namespace Alpha {
-  export {
-    Inference as Inference,
-    type InferenceRerankResponse as InferenceRerankResponse,
-    type InferenceRerankParams as InferenceRerankParams,
-  };
-
   export {
     PostTraining as PostTraining,
     type AlgorithmConfig as AlgorithmConfig,
+    type ListPostTrainingJobsResponse as ListPostTrainingJobsResponse,
     type PostTrainingJob as PostTrainingJob,
     type PostTrainingPreferenceOptimizeParams as PostTrainingPreferenceOptimizeParams,
     type PostTrainingSupervisedFineTuneParams as PostTrainingSupervisedFineTuneParams,
@@ -82,4 +82,12 @@ export declare namespace Alpha {
     type EvalRunEvalParams as EvalRunEvalParams,
     type EvalRunEvalAlphaParams as EvalRunEvalAlphaParams,
   };
+
+  export { Admin as Admin, type AdminListRoutesParams as AdminListRoutesParams };
+
+  export {
+    Inference as Inference,
+    type InferenceRerankResponse as InferenceRerankResponse,
+    type InferenceRerankParams as InferenceRerankParams,
+  };
 }

package/src/resources/alpha/benchmarks.ts

@@ -59,6 +59,9 @@ export class Benchmarks extends APIResource {
  * A benchmark resource for evaluating model performance.
  */
 export interface Benchmark {
+  /**
+   * Identifier of the dataset to use for the benchmark evaluation.
+   */
   dataset_id: string;
 
   /**
@@ -71,10 +74,13 @@ export interface Benchmark {
    */
   provider_id: string;
 
+  /**
+   * List of scoring function identifiers to apply during evaluation.
+   */
   scoring_functions: Array<string>;
 
   /**
-   * Metadata for this evaluation task
+   * Metadata for this evaluation task.
    */
   metadata?: { [key: string]: unknown };
 
@@ -83,26 +89,56 @@ export interface Benchmark {
    */
   provider_resource_id?: string | null;
 
+  /**
+   * The resource type, always benchmark.
+   */
   type?: 'benchmark';
 }
 
+/**
+ * Response containing a list of benchmark objects.
+ */
 export interface ListBenchmarksResponse {
+  /**
+   * List of benchmark objects.
+   */
   data: BenchmarkListResponse;
 }
 
+/**
+ * List of benchmark objects.
+ */
 export type BenchmarkListResponse = Array<Benchmark>;
 
 export interface BenchmarkRegisterParams {
+  /**
+   * The ID of the benchmark to register.
+   */
   benchmark_id: string;
 
+  /**
+   * The ID of the dataset to use for the benchmark.
+   */
   dataset_id: string;
 
+  /**
+   * The scoring functions to use for the benchmark.
+   */
   scoring_functions: Array<string>;
 
+  /**
+   * The metadata to use for the benchmark.
+   */
   metadata?: { [key: string]: unknown } | null;
 
+  /**
+   * The ID of the provider benchmark to use for the benchmark.
+   */
   provider_benchmark_id?: string | null;
 
+  /**
+   * The ID of the provider to use for the benchmark.
+   */
   provider_id?: string | null;
 }