llama-stack-client 0.2.10 → 0.2.11-rc1

package/resources/vector-stores.mjs

@@ -0,0 +1,3 @@
+// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+export * from "./vector-stores/index.mjs";
+//# sourceMappingURL=vector-stores.mjs.map

package/resources/vector-stores.mjs.map

@@ -0,0 +1 @@
+{"version":3,"file":"vector-stores.mjs","sourceRoot":"","sources":["../src/resources/vector-stores.ts"],"names":[],"mappings":"AAAA,sFAAsF"}

package/src/index.ts

@@ -32,6 +32,16 @@ import {
   Datasets,
   ListDatasetsResponse,
 } from './resources/datasets';
+import { CreateEmbeddingsResponse, EmbeddingCreateParams, Embeddings } from './resources/embeddings';
+import {
+  DeleteFileResponse,
+  File,
+  FileContentResponse,
+  FileCreateParams,
+  FileListParams,
+  Files,
+  ListFilesResponse,
+} from './resources/files';
 import {
   ChatCompletionResponseStreamChunk,
   CompletionResponse,
@@ -174,6 +184,17 @@ import {
   ToolRuntimeListToolsParams,
   ToolRuntimeListToolsResponse,
 } from './resources/tool-runtime/tool-runtime';
+import {
+  ListVectorStoresResponse,
+  VectorStore,
+  VectorStoreCreateParams,
+  VectorStoreDeleteResponse,
+  VectorStoreListParams,
+  VectorStoreSearchParams,
+  VectorStoreSearchResponse,
+  VectorStoreUpdateParams,
+  VectorStores,
+} from './resources/vector-stores/vector-stores';
 
 export interface ClientOptions {
   /**
@@ -291,10 +312,12 @@ export class LlamaStackClient extends Core.APIClient {
   eval: API.Eval = new API.Eval(this);
   inspect: API.Inspect = new API.Inspect(this);
   inference: API.Inference = new API.Inference(this);
+  embeddings: API.Embeddings = new API.Embeddings(this);
   chat: API.Chat = new API.Chat(this);
   completions: API.Completions = new API.Completions(this);
   vectorIo: API.VectorIo = new API.VectorIo(this);
   vectorDBs: API.VectorDBs = new API.VectorDBs(this);
+  vectorStores: API.VectorStores = new API.VectorStores(this);
   models: API.Models = new API.Models(this);
   postTraining: API.PostTraining = new API.PostTraining(this);
   providers: API.Providers = new API.Providers(this);
@@ -306,6 +329,7 @@ export class LlamaStackClient extends Core.APIClient {
   scoring: API.Scoring = new API.Scoring(this);
   scoringFunctions: API.ScoringFunctions = new API.ScoringFunctions(this);
   benchmarks: API.Benchmarks = new API.Benchmarks(this);
+  files: API.Files = new API.Files(this);
 
   protected override defaultQuery(): Core.DefaultQuery | undefined {
     return this._options.defaultQuery;
@@ -359,10 +383,12 @@ LlamaStackClient.Datasets = Datasets;
 LlamaStackClient.Eval = Eval;
 LlamaStackClient.Inspect = Inspect;
 LlamaStackClient.Inference = Inference;
+LlamaStackClient.Embeddings = Embeddings;
 LlamaStackClient.Chat = Chat;
 LlamaStackClient.Completions = Completions;
 LlamaStackClient.VectorIo = VectorIo;
 LlamaStackClient.VectorDBs = VectorDBs;
+LlamaStackClient.VectorStores = VectorStores;
 LlamaStackClient.Models = Models;
 LlamaStackClient.PostTraining = PostTraining;
 LlamaStackClient.Providers = Providers;
@@ -374,6 +400,7 @@ LlamaStackClient.Telemetry = Telemetry;
 LlamaStackClient.Scoring = Scoring;
 LlamaStackClient.ScoringFunctions = ScoringFunctions;
 LlamaStackClient.Benchmarks = Benchmarks;
+LlamaStackClient.Files = Files;
 export declare namespace LlamaStackClient {
   export type RequestOptions = Core.RequestOptions;
 
@@ -479,6 +506,12 @@ export declare namespace LlamaStackClient {
     type InferenceEmbeddingsParams as InferenceEmbeddingsParams,
   };
 
+  export {
+    Embeddings as Embeddings,
+    type CreateEmbeddingsResponse as CreateEmbeddingsResponse,
+    type EmbeddingCreateParams as EmbeddingCreateParams,
+  };
+
   export { Chat as Chat, type ChatCompletionChunk as ChatCompletionChunk };
 
   export {
@@ -505,6 +538,18 @@ export declare namespace LlamaStackClient {
     type VectorDBRegisterParams as VectorDBRegisterParams,
   };
 
+  export {
+    VectorStores as VectorStores,
+    type ListVectorStoresResponse as ListVectorStoresResponse,
+    type VectorStore as VectorStore,
+    type VectorStoreDeleteResponse as VectorStoreDeleteResponse,
+    type VectorStoreSearchResponse as VectorStoreSearchResponse,
+    type VectorStoreCreateParams as VectorStoreCreateParams,
+    type VectorStoreUpdateParams as VectorStoreUpdateParams,
+    type VectorStoreListParams as VectorStoreListParams,
+    type VectorStoreSearchParams as VectorStoreSearchParams,
+  };
+
   export {
     Models as Models,
     type ListModelsResponse as ListModelsResponse,
@@ -597,6 +642,16 @@ export declare namespace LlamaStackClient {
     type BenchmarkRegisterParams as BenchmarkRegisterParams,
   };
 
+  export {
+    Files as Files,
+    type DeleteFileResponse as DeleteFileResponse,
+    type File as File,
+    type ListFilesResponse as ListFilesResponse,
+    type FileContentResponse as FileContentResponse,
+    type FileCreateParams as FileCreateParams,
+    type FileListParams as FileListParams,
+  };
+
   export type AgentConfig = API.AgentConfig;
   export type BatchCompletion = API.BatchCompletion;
   export type ChatCompletionResponse = API.ChatCompletionResponse;

package/src/resource.ts

@@ -2,7 +2,7 @@
 
 import type { LlamaStackClient } from './index';
 
-export class APIResource {
+export abstract class APIResource {
   protected _client: LlamaStackClient;
 
   constructor(client: LlamaStackClient) {

package/src/resources/completions.ts

@@ -223,6 +223,11 @@ export interface CompletionCreateParamsBase {
    */
   stream_options?: Record<string, boolean | number | string | Array<unknown> | unknown | null>;
 
+  /**
+   * (Optional) The suffix that should be appended to the completion.
+   */
+  suffix?: string;
+
   /**
    * (Optional) The temperature to use.
    */

package/src/resources/datasets.ts

@@ -195,6 +195,11 @@ export interface DatasetIterrowsResponse {
    * Whether there are more items available after this set
    */
   has_more: boolean;
+
+  /**
+   * The URL for accessing this list
+   */
+  url?: string;
 }
 
 export interface DatasetRegisterResponse {

package/src/resources/embeddings.ts

@@ -0,0 +1,119 @@
+// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+import { APIResource } from '../resource';
+import * as Core from '../core';
+
+export class Embeddings extends APIResource {
+  /**
+   * Generate OpenAI-compatible embeddings for the given input using the specified
+   * model.
+   */
+  create(
+    body: EmbeddingCreateParams,
+    options?: Core.RequestOptions,
+  ): Core.APIPromise<CreateEmbeddingsResponse> {
+    return this._client.post('/v1/openai/v1/embeddings', { body, ...options });
+  }
+}
+
+/**
+ * Response from an OpenAI-compatible embeddings request.
+ */
+export interface CreateEmbeddingsResponse {
+  /**
+   * List of embedding data objects
+   */
+  data: Array<CreateEmbeddingsResponse.Data>;
+
+  /**
+   * The model that was used to generate the embeddings
+   */
+  model: string;
+
+  /**
+   * The object type, which will be "list"
+   */
+  object: 'list';
+
+  /**
+   * Usage information
+   */
+  usage: CreateEmbeddingsResponse.Usage;
+}
+
+export namespace CreateEmbeddingsResponse {
+  /**
+   * A single embedding data object from an OpenAI-compatible embeddings response.
+   */
+  export interface Data {
+    /**
+     * The embedding vector as a list of floats (when encoding_format="float") or as a
+     * base64-encoded string (when encoding_format="base64")
+     */
+    embedding: Array<number> | string;
+
+    /**
+     * The index of the embedding in the input list
+     */
+    index: number;
+
+    /**
+     * The object type, which will be "embedding"
+     */
+    object: 'embedding';
+  }
+
+  /**
+   * Usage information
+   */
+  export interface Usage {
+    /**
+     * The number of tokens in the input
+     */
+    prompt_tokens: number;
+
+    /**
+     * The total number of tokens used
+     */
+    total_tokens: number;
+  }
+}
+
+export interface EmbeddingCreateParams {
+  /**
+   * Input text to embed, encoded as a string or array of strings. To embed multiple
+   * inputs in a single request, pass an array of strings.
+   */
+  input: string | Array<string>;
+
+  /**
+   * The identifier of the model to use. The model must be an embedding model
+   * registered with Llama Stack and available via the /models endpoint.
+   */
+  model: string;
+
+  /**
+   * (Optional) The number of dimensions the resulting output embeddings should have.
+   * Only supported in text-embedding-3 and later models.
+   */
+  dimensions?: number;
+
+  /**
+   * (Optional) The format to return the embeddings in. Can be either "float" or
+   * "base64". Defaults to "float".
+   */
+  encoding_format?: string;
+
+  /**
+   * (Optional) A unique identifier representing your end-user, which can help OpenAI
+   * to monitor and detect abuse.
+   */
+  user?: string;
+}
+
+export declare namespace Embeddings {
+  export {
+    type CreateEmbeddingsResponse as CreateEmbeddingsResponse,
+    type EmbeddingCreateParams as EmbeddingCreateParams,
+  };
+}

package/src/resources/files.ts

@@ -0,0 +1,184 @@
+// 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';
+
+export class Files extends APIResource {
+  /**
+   * Upload a file that can be used across various endpoints. The file upload should
+   * be a multipart form request with:
+   *
+   * - file: The File object (not file name) to be uploaded.
+   * - purpose: The intended purpose of the uploaded file.
+   */
+  create(body: FileCreateParams, options?: Core.RequestOptions): Core.APIPromise<File> {
+    return this._client.post('/v1/openai/v1/files', Core.multipartFormRequestOptions({ body, ...options }));
+  }
+
+  /**
+   * Returns information about a specific file.
+   */
+  retrieve(fileId: string, options?: Core.RequestOptions): Core.APIPromise<File> {
+    return this._client.get(`/v1/openai/v1/files/${fileId}`, options);
+  }
+
+  /**
+   * Returns a list of files that belong to the user's organization.
+   */
+  list(query?: FileListParams, options?: Core.RequestOptions): Core.APIPromise<ListFilesResponse>;
+  list(options?: Core.RequestOptions): Core.APIPromise<ListFilesResponse>;
+  list(
+    query: FileListParams | Core.RequestOptions = {},
+    options?: Core.RequestOptions,
+  ): Core.APIPromise<ListFilesResponse> {
+    if (isRequestOptions(query)) {
+      return this.list({}, query);
+    }
+    return this._client.get('/v1/openai/v1/files', { query, ...options });
+  }
+
+  /**
+   * Delete a file.
+   */
+  delete(fileId: string, options?: Core.RequestOptions): Core.APIPromise<DeleteFileResponse> {
+    return this._client.delete(`/v1/openai/v1/files/${fileId}`, options);
+  }
+
+  /**
+   * Returns the contents of the specified file.
+   */
+  content(fileId: string, options?: Core.RequestOptions): Core.APIPromise<unknown> {
+    return this._client.get(`/v1/openai/v1/files/${fileId}/content`, options);
+  }
+}
+
+/**
+ * Response for deleting a file in OpenAI Files API.
+ */
+export interface DeleteFileResponse {
+  /**
+   * The file identifier that was deleted
+   */
+  id: string;
+
+  /**
+   * Whether the file was successfully deleted
+   */
+  deleted: boolean;
+
+  /**
+   * The object type, which is always "file"
+   */
+  object: 'file';
+}
+
+/**
+ * OpenAI File object as defined in the OpenAI Files API.
+ */
+export interface File {
+  /**
+   * The file identifier, which can be referenced in the API endpoints
+   */
+  id: string;
+
+  /**
+   * The size of the file, in bytes
+   */
+  bytes: number;
+
+  /**
+   * The Unix timestamp (in seconds) for when the file was created
+   */
+  created_at: number;
+
+  /**
+   * The Unix timestamp (in seconds) for when the file expires
+   */
+  expires_at: number;
+
+  /**
+   * The name of the file
+   */
+  filename: string;
+
+  /**
+   * The object type, which is always "file"
+   */
+  object: 'file';
+
+  /**
+   * The intended purpose of the file
+   */
+  purpose: 'assistants';
+}
+
+/**
+ * Response for listing files in OpenAI Files API.
+ */
+export interface ListFilesResponse {
+  /**
+   * List of file objects
+   */
+  data: Array<File>;
+
+  first_id: string;
+
+  has_more: boolean;
+
+  last_id: string;
+
+  /**
+   * The object type, which is always "list"
+   */
+  object: 'list';
+}
+
+export type FileContentResponse = unknown;
+
+export interface FileCreateParams {
+  file: Core.Uploadable;
+
+  /**
+   * Valid purpose values for OpenAI Files API.
+   */
+  purpose: 'assistants';
+}
+
+export interface FileListParams {
+  /**
+   * A cursor for use in pagination. `after` is an object ID that defines your place
+   * in the list. For instance, if you make a list request and receive 100 objects,
+   * ending with obj_foo, your subsequent call can include after=obj_foo in order to
+   * fetch the next page of the list.
+   */
+  after?: string;
+
+  /**
+   * A limit on the number of objects to be returned. Limit can range between 1 and
+   * 10,000, and the default is 10,000.
+   */
+  limit?: number;
+
+  /**
+   * Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+   * order and `desc` for descending order.
+   */
+  order?: 'asc' | 'desc';
+
+  /**
+   * Only return files with the given purpose.
+   */
+  purpose?: 'assistants';
+}
+
+export declare namespace Files {
+  export {
+    type DeleteFileResponse as DeleteFileResponse,
+    type File as File,
+    type ListFilesResponse as ListFilesResponse,
+    type FileContentResponse as FileContentResponse,
+    type FileCreateParams as FileCreateParams,
+    type FileListParams as FileListParams,
+  };
+}

package/src/resources/index.ts

@@ -36,6 +36,7 @@ export {
   type DatasetIterrowsParams,
   type DatasetRegisterParams,
 } from './datasets';
+export { Embeddings, type CreateEmbeddingsResponse, type EmbeddingCreateParams } from './embeddings';
 export {
   Eval,
   type BenchmarkConfig,
@@ -47,6 +48,15 @@ export {
   type EvalRunEvalParams,
   type EvalRunEvalAlphaParams,
 } from './eval/eval';
+export {
+  Files,
+  type DeleteFileResponse,
+  type File,
+  type ListFilesResponse,
+  type FileContentResponse,
+  type FileCreateParams,
+  type FileListParams,
+} from './files';
 export {
   Inference,
   type ChatCompletionResponseStreamChunk,
@@ -173,3 +183,14 @@ export {
   type VectorIoInsertParams,
   type VectorIoQueryParams,
 } from './vector-io';
+export {
+  VectorStores,
+  type ListVectorStoresResponse,
+  type VectorStore,
+  type VectorStoreDeleteResponse,
+  type VectorStoreSearchResponse,
+  type VectorStoreCreateParams,
+  type VectorStoreUpdateParams,
+  type VectorStoreListParams,
+  type VectorStoreSearchParams,
+} from './vector-stores/vector-stores';

package/src/resources/responses/input-items.ts

@@ -29,6 +29,7 @@ export class InputItems extends APIResource {
 export interface InputItemListResponse {
   data: Array<
     | InputItemListResponse.OpenAIResponseOutputMessageWebSearchToolCall
+    | InputItemListResponse.OpenAIResponseOutputMessageFileSearchToolCall
     | InputItemListResponse.OpenAIResponseOutputMessageFunctionToolCall
     | InputItemListResponse.OpenAIResponseInputFunctionToolCallOutput
     | InputItemListResponse.OpenAIResponseMessage
@@ -46,6 +47,18 @@ export namespace InputItemListResponse {
     type: 'web_search_call';
   }
 
+  export interface OpenAIResponseOutputMessageFileSearchToolCall {
+    id: string;
+
+    queries: Array<string>;
+
+    status: string;
+
+    type: 'file_search_call';
+
+    results?: Array<Record<string, boolean | number | string | Array<unknown> | unknown | null>>;
+  }
+
   export interface OpenAIResponseOutputMessageFunctionToolCall {
     arguments: string;