feat(api): sync

Author: stainless-app[bot]

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 104
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/llamastack%2Fllama-stack-client-2acd62f8d5c4523bad4ddb2cc50608135249858b7047a71b48e25befa1e1f43f.yml
-openapi_spec_hash: 1ad726ff81dc21720c8c3443d33c0562
-config_hash: 734f75f2a6b46155e1852d378118b5e8
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/llamastack%2Fllama-stack-client-fcaa960dc2de2029f4f67f13ff1d0fc1ff70e683810ed9739be805debef1673d.yml
+openapi_spec_hash: a25e7616ad6230f872b46c2cb6fa0a96
+config_hash: ff421daf28f90ad4bd4e13f374b18a00

README.md

@@ -27,11 +27,33 @@ import LlamaStackClient from 'llama-stack-client';
 
 const client = new LlamaStackClient();
 
-const healthInfo = await client.inspect.health();
+const model = await client.models.register({ model_id: 'model_id' });
 
-console.log(healthInfo.status);
+console.log(model.identifier);
 ```
 
+## Streaming responses
+
+We provide support for streaming responses using Server Sent Events (SSE).
+
+```ts
+import LlamaStackClient from 'llama-stack-client';
+
+const client = new LlamaStackClient();
+
+const stream = await client.chat.completions.create({
+  messages: [{ content: 'string', role: 'user' }],
+  model: 'model',
+  stream: true,
+});
+for await (const chatCompletionChunk of stream) {
+  console.log(chatCompletionChunk);
+}
+```
+
+If you need to cancel a stream, you can `break` from the loop
+or call `stream.controller.abort()`.
+
 ### Request & Response types
 
 This library includes TypeScript definitions for all request params and response fields. You may import and use them like so:
@@ -42,7 +64,13 @@ import LlamaStackClient from 'llama-stack-client';
 
 const client = new LlamaStackClient();
 
-const healthInfo: LlamaStackClient.HealthInfo = await client.inspect.health();
+const params: LlamaStackClient.Chat.CompletionCreateParams = {
+  messages: [{ content: 'string', role: 'user' }],
+  model: 'model',
+};
+const completion: LlamaStackClient.Chat.CompletionCreateResponse = await client.chat.completions.create(
+  params,
+);
 ```
 
 Documentation for each method, request param, and response field are available in docstrings and will appear on hover in most modern editors.
@@ -85,15 +113,17 @@ a subclass of `APIError` will be thrown:
 
 <!-- prettier-ignore -->
 ```ts
-const healthInfo = await client.inspect.health().catch(async (err) => {
-  if (err instanceof LlamaStackClient.APIError) {
-    console.log(err.status); // 400
-    console.log(err.name); // BadRequestError
-    console.log(err.headers); // {server: 'nginx', ...}
-  } else {
-    throw err;
-  }
-});
+const completion = await client.chat.completions
+  .create({ messages: [{ content: 'string', role: 'user' }], model: 'model' })
+  .catch(async (err) => {
+    if (err instanceof LlamaStackClient.APIError) {
+      console.log(err.status); // 400
+      console.log(err.name); // BadRequestError
+      console.log(err.headers); // {server: 'nginx', ...}
+    } else {
+      throw err;
+    }
+  });
 ```
 
 Error codes are as follows:
@@ -125,7 +155,7 @@ const client = new LlamaStackClient({
 });
 
 // Or, configure per-request:
-await client.inspect.health({
+await client.chat.completions.create({ messages: [{ content: 'string', role: 'user' }], model: 'model' }, {
   maxRetries: 5,
 });
 ```
@@ -142,7 +172,7 @@ const client = new LlamaStackClient({
 });
 
 // Override per-request:
-await client.inspect.health({
+await client.chat.completions.create({ messages: [{ content: 'string', role: 'user' }], model: 'model' }, {
   timeout: 5 * 1000,
 });
 ```
@@ -163,13 +193,17 @@ You can also use the `.withResponse()` method to get the raw `Response` along wi
 ```ts
 const client = new LlamaStackClient();
 
-const response = await client.inspect.health().asResponse();
+const response = await client.chat.completions
+  .create({ messages: [{ content: 'string', role: 'user' }], model: 'model' })
+  .asResponse();
 console.log(response.headers.get('X-My-Header'));
 console.log(response.statusText); // access the underlying Response object
 
-const { data: healthInfo, response: raw } = await client.inspect.health().withResponse();
+const { data: completion, response: raw } = await client.chat.completions
+  .create({ messages: [{ content: 'string', role: 'user' }], model: 'model' })
+  .withResponse();
 console.log(raw.headers.get('X-My-Header'));
-console.log(healthInfo.status);
+console.log(completion);
 ```
 
 ### Making custom/undocumented requests
@@ -273,9 +307,12 @@ const client = new LlamaStackClient({
 });
 
 // Override per-request:
-await client.inspect.health({
-  httpAgent: new http.Agent({ keepAlive: false }),
-});
+await client.chat.completions.create(
+  { messages: [{ content: 'string', role: 'user' }], model: 'model' },
+  {
+    httpAgent: new http.Agent({ keepAlive: false }),
+  },
+);
 ```
 
 ## Semantic versioning

api.md

@@ -323,6 +323,8 @@ Methods:
 
 - <code title="post /v1/synthetic-data-generation/generate">client.syntheticDataGeneration.<a href="./src/resources/synthetic-data-generation.ts">generate</a>({ ...params }) -> SyntheticDataGenerationResponse</code>
 
+# Telemetry
+
 # Scoring
 
 Types:
@@ -381,27 +383,6 @@ Methods:
 - <code title="delete /v1/files/{file_id}">client.files.<a href="./src/resources/files.ts">delete</a>(fileId) -> DeleteFileResponse</code>
 - <code title="get /v1/files/{file_id}/content">client.files.<a href="./src/resources/files.ts">content</a>(fileId) -> unknown</code>
 
-# Beta
-
-## Datasets
-
-Types:
-
-- <code><a href="./src/resources/beta/datasets.ts">ListDatasetsResponse</a></code>
-- <code><a href="./src/resources/beta/datasets.ts">DatasetRetrieveResponse</a></code>
-- <code><a href="./src/resources/beta/datasets.ts">DatasetListResponse</a></code>
-- <code><a href="./src/resources/beta/datasets.ts">DatasetIterrowsResponse</a></code>
-- <code><a href="./src/resources/beta/datasets.ts">DatasetRegisterResponse</a></code>
-
-Methods:
-
-- <code title="get /v1beta/datasets/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">retrieve</a>(datasetId) -> DatasetRetrieveResponse</code>
-- <code title="get /v1beta/datasets">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">list</a>() -> DatasetListResponse</code>
-- <code title="post /v1beta/datasetio/append-rows/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">appendrows</a>(datasetId, { ...params }) -> void</code>
-- <code title="get /v1beta/datasetio/iterrows/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">iterrows</a>(datasetId, { ...params }) -> DatasetIterrowsResponse</code>
-- <code title="post /v1beta/datasets">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">register</a>({ ...params }) -> DatasetRegisterResponse</code>
-- <code title="delete /v1beta/datasets/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">unregister</a>(datasetId) -> void</code>
-
 # Alpha
 
 ## Inference
@@ -523,3 +504,24 @@ Methods:
 - <code title="post /v1alpha/agents/{agent_id}/session/{session_id}/turn">client.alpha.agents.turn.<a href="./src/resources/alpha/agents/turn.ts">create</a>(agentId, sessionId, { ...params }) -> Turn</code>
 - <code title="get /v1alpha/agents/{agent_id}/session/{session_id}/turn/{turn_id}">client.alpha.agents.turn.<a href="./src/resources/alpha/agents/turn.ts">retrieve</a>(agentId, sessionId, turnId) -> Turn</code>
 - <code title="post /v1alpha/agents/{agent_id}/session/{session_id}/turn/{turn_id}/resume">client.alpha.agents.turn.<a href="./src/resources/alpha/agents/turn.ts">resume</a>(agentId, sessionId, turnId, { ...params }) -> Turn</code>
+
+# Beta
+
+## Datasets
+
+Types:
+
+- <code><a href="./src/resources/beta/datasets.ts">ListDatasetsResponse</a></code>
+- <code><a href="./src/resources/beta/datasets.ts">DatasetRetrieveResponse</a></code>
+- <code><a href="./src/resources/beta/datasets.ts">DatasetListResponse</a></code>
+- <code><a href="./src/resources/beta/datasets.ts">DatasetIterrowsResponse</a></code>
+- <code><a href="./src/resources/beta/datasets.ts">DatasetRegisterResponse</a></code>
+
+Methods:
+
+- <code title="get /v1beta/datasets/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">retrieve</a>(datasetId) -> DatasetRetrieveResponse</code>
+- <code title="get /v1beta/datasets">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">list</a>() -> DatasetListResponse</code>
+- <code title="post /v1beta/datasetio/append-rows/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">appendrows</a>(datasetId, { ...params }) -> void</code>
+- <code title="get /v1beta/datasetio/iterrows/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">iterrows</a>(datasetId, { ...params }) -> DatasetIterrowsResponse</code>
+- <code title="post /v1beta/datasets">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">register</a>({ ...params }) -> DatasetRegisterResponse</code>
+- <code title="delete /v1beta/datasets/{dataset_id}">client.beta.datasets.<a href="./src/resources/beta/datasets.ts">unregister</a>(datasetId) -> void</code>

src/index.ts

@@ -70,6 +70,7 @@ import {
   SyntheticDataGenerationGenerateParams,
   SyntheticDataGenerationResponse,
 } from './resources/synthetic-data-generation';
+import { Telemetry } from './resources/telemetry';
 import {
   ListToolGroupsResponse,
   ToolGroup,
@@ -262,12 +263,13 @@ export class LlamaStackClient extends Core.APIClient {
   safety: API.Safety = new API.Safety(this);
   shields: API.Shields = new API.Shields(this);
   syntheticDataGeneration: API.SyntheticDataGeneration = new API.SyntheticDataGeneration(this);
+  telemetry: API.Telemetry = new API.Telemetry(this);
   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);
-  beta: API.Beta = new API.Beta(this);
   alpha: API.Alpha = new API.Alpha(this);
+  beta: API.Beta = new API.Beta(this);
 
   /**
    * Check whether the base URL is set to its default.
@@ -339,13 +341,14 @@ LlamaStackClient.Moderations = Moderations;
 LlamaStackClient.Safety = Safety;
 LlamaStackClient.Shields = Shields;
 LlamaStackClient.SyntheticDataGeneration = SyntheticDataGeneration;
+LlamaStackClient.Telemetry = Telemetry;
 LlamaStackClient.Scoring = Scoring;
 LlamaStackClient.ScoringFunctions = ScoringFunctions;
 LlamaStackClient.Benchmarks = Benchmarks;
 LlamaStackClient.Files = Files;
 LlamaStackClient.FilesOpenAICursorPage = FilesOpenAICursorPage;
-LlamaStackClient.Beta = Beta;
 LlamaStackClient.Alpha = Alpha;
+LlamaStackClient.Beta = Beta;
 
 export declare namespace LlamaStackClient {
   export type RequestOptions = Core.RequestOptions;
@@ -492,6 +495,8 @@ export declare namespace LlamaStackClient {
     type SyntheticDataGenerationGenerateParams as SyntheticDataGenerationGenerateParams,
   };
 
+  export { Telemetry as Telemetry };
+
   export {
     Scoring as Scoring,
     type ScoringScoreResponse as ScoringScoreResponse,
@@ -528,10 +533,10 @@ export declare namespace LlamaStackClient {
     type FileListParams as FileListParams,
   };
 
-  export { Beta as Beta };
-
   export { Alpha as Alpha };
 
+  export { Beta as Beta };
+
   export type AgentConfig = API.AgentConfig;
   export type CompletionMessage = API.CompletionMessage;
   export type Document = API.Document;

src/resources/alpha/agents/agents.ts

@@ -137,7 +137,7 @@ export interface MemoryRetrievalStep {
   /**
    * The IDs of the vector databases to retrieve context from.
    */
-  vector_store_ids: string;
+  vector_db_ids: string;
 
   /**
    * The time the step completed.

src/resources/index.ts

@@ -87,6 +87,7 @@ export {
   type SyntheticDataGenerationResponse,
   type SyntheticDataGenerationGenerateParams,
 } from './synthetic-data-generation';
+export { Telemetry } from './telemetry';
 export {
   ToolRuntime,
   type ToolDef,

src/resources/telemetry.ts

@@ -0,0 +1,5 @@
+// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+import { APIResource } from '../resource';
+
+export class Telemetry extends APIResource {}

src/resources/tool-runtime/rag-tool.ts

@@ -38,7 +38,7 @@ export interface RagToolInsertParams {
   /**
    * ID of the vector database to store the document embeddings
    */
-  vector_store_id: string;
+  vector_db_id: string;
 }
 
 export interface RagToolQueryParams {
@@ -50,7 +50,7 @@ export interface RagToolQueryParams {
   /**
    * List of vector database IDs to search within
    */
-  vector_store_ids: Array<string>;
+  vector_db_ids: Array<string>;
 
   /**
    * (Optional) Configuration parameters for the query operation

src/resources/vector-io.ts

@@ -151,7 +151,7 @@ export interface VectorIoInsertParams {
   /**
    * The identifier of the vector database to insert the chunks into.
    */
-  vector_store_id: string;
+  vector_db_id: string;
 
   /**
    * The time to live of the chunks.
@@ -267,7 +267,7 @@ export interface VectorIoQueryParams {
   /**
    * The identifier of the vector database to query.
    */
-  vector_store_id: string;
+  vector_db_id: string;
 
   /**
    * The parameters of the query.

tests/api-resources/tool-runtime/rag-tool.test.ts

@@ -10,7 +10,7 @@ describe('resource ragTool', () => {
     const responsePromise = client.toolRuntime.ragTool.insert({
       chunk_size_in_tokens: 0,
       documents: [{ content: 'string', document_id: 'document_id', metadata: { foo: true } }],
-      vector_store_id: 'vector_store_id',
+      vector_db_id: 'vector_db_id',
     });
     const rawResponse = await responsePromise.asResponse();
     expect(rawResponse).toBeInstanceOf(Response);
@@ -27,14 +27,14 @@ describe('resource ragTool', () => {
       documents: [
         { content: 'string', document_id: 'document_id', metadata: { foo: true }, mime_type: 'mime_type' },
       ],
-      vector_store_id: 'vector_store_id',
+      vector_db_id: 'vector_db_id',
     });
   });
 
   test('query: only required params', async () => {
     const responsePromise = client.toolRuntime.ragTool.query({
       content: 'string',
-      vector_store_ids: ['string'],
+      vector_db_ids: ['string'],
     });
     const rawResponse = await responsePromise.asResponse();
     expect(rawResponse).toBeInstanceOf(Response);
@@ -48,7 +48,7 @@ describe('resource ragTool', () => {
   test('query: required and optional params', async () => {
     const response = await client.toolRuntime.ragTool.query({
       content: 'string',
-      vector_store_ids: ['string'],
+      vector_db_ids: ['string'],
       query_config: {
         chunk_template: 'chunk_template',
         max_chunks: 0,