llama-stack-client 0.6.1 → 0.7.1

package/dist/src/resources/alpha/eval/jobs.ts

@@ -10,6 +10,9 @@ import { APIResource } from '../../../resource';
 import * as Core from '../../../core';
 import * as EvalAPI from './eval';
 
+/**
+ * Llama Stack Evaluation API for running evaluations on model and agent candidates.
+ */
 export class Jobs extends APIResource {
   /**
    * Get the result of a job.

package/dist/src/resources/alpha/index.ts

@@ -26,11 +26,3 @@ export {
   type EvalRunEvalAlphaParams,
 } from './eval/index';
 export { Inference, type InferenceRerankResponse, type InferenceRerankParams } from './inference';
-export {
-  PostTraining,
-  type AlgorithmConfig,
-  type ListPostTrainingJobsResponse,
-  type PostTrainingJob,
-  type PostTrainingPreferenceOptimizeParams,
-  type PostTrainingSupervisedFineTuneParams,
-} from './post-training/index';

package/dist/src/resources/alpha/inference.ts

@@ -9,6 +9,14 @@
 import { APIResource } from '../../resource';
 import * as Core from '../../core';
 
+/**
+ * Llama Stack Inference API for generating completions, chat completions, and embeddings.
+ *
+ * This API provides the raw interface to the underlying models. Three kinds of models are supported:
+ * - LLM models: these models generate "raw" and "chat" (conversational) completions.
+ * - Embedding models: these models generate embeddings to be used for semantic search.
+ * - Rerank models: these models reorder the documents based on their relevance to a query.
+ */
 export class Inference extends APIResource {
   /**
    * Rerank a list of documents based on their relevance to a query.

package/dist/src/resources/batches.ts

@@ -11,6 +11,14 @@ import { isRequestOptions } from '../core';
 import * as Core from '../core';
 import { OpenAICursorPage, type OpenAICursorPageParams } from '../pagination';
 
+/**
+ * The API is designed to allow use of openai client libraries for seamless integration.
+ *
+ * This API provides the following extensions:
+ *  - idempotent batch creation
+ *
+ * Note: This API is currently under active development and may undergo changes.
+ */
 export class Batches extends APIResource {
   /**
    * Create a new batch for processing multiple API requests.

package/dist/src/resources/beta/datasets.ts

@@ -124,7 +124,7 @@ export interface DatasetRetrieveResponse {
   /**
    * Purpose of the dataset indicating its intended use
    */
-  purpose: 'post-training/messages' | 'eval/question-answer' | 'eval/messages-answer';
+  purpose: 'eval/question-answer' | 'eval/messages-answer';
 
   /**
    * Data source configuration for the dataset
@@ -205,7 +205,7 @@ export namespace DatasetListResponse {
     /**
      * Purpose of the dataset indicating its intended use
      */
-    purpose: 'post-training/messages' | 'eval/question-answer' | 'eval/messages-answer';
+    purpose: 'eval/question-answer' | 'eval/messages-answer';
 
     /**
      * Data source configuration for the dataset
@@ -292,7 +292,7 @@ export interface DatasetRegisterResponse {
   /**
    * Purpose of the dataset indicating its intended use
    */
-  purpose: 'post-training/messages' | 'eval/question-answer' | 'eval/messages-answer';
+  purpose: 'eval/question-answer' | 'eval/messages-answer';
 
   /**
    * Data source configuration for the dataset
@@ -351,11 +351,6 @@ export namespace DatasetRegisterResponse {
 }
 
 export interface DatasetAppendrowsParams {
-  /**
-   * The ID of the dataset to append the rows to.
-   */
-  body_dataset_id: string;
-
   /**
    * The rows to append to the dataset.
    */
@@ -378,7 +373,7 @@ export interface DatasetRegisterParams {
   /**
    * The purpose of the dataset.
    */
-  purpose: 'post-training/messages' | 'eval/question-answer' | 'eval/messages-answer';
+  purpose: 'eval/question-answer' | 'eval/messages-answer';
 
   /**
    * The data source of the dataset.

package/dist/src/resources/chat/chat.ts

@@ -53,7 +53,17 @@ export interface ChatCompletionChunk {
   object?: 'chat.completion.chunk';
 
   /**
-   * Usage information for OpenAI chat completion.
+   * The service tier that was used for this response.
+   */
+  service_tier?: string | null;
+
+  /**
+   * System fingerprint for this completion chunk.
+   */
+  system_fingerprint?: string | null;
+
+  /**
+   * Token usage information (typically included in final chunk with stream_options).
    */
   usage?: ChatCompletionChunk.Usage | null;
 }
@@ -79,10 +89,9 @@ export namespace ChatCompletionChunk {
     finish_reason?: 'stop' | 'length' | 'tool_calls' | 'content_filter' | 'function_call' | null;
 
     /**
-     * The log probabilities for the tokens in the message from an OpenAI-compatible
-     * chat completion response.
+     * The log probabilities for the tokens in the message.
      */
-    logprobs?: Choice.Logprobs | null;
+    logprobs?: Choice.Logprobs;
   }
 
   export namespace Choice {
@@ -108,7 +117,7 @@ export namespace ChatCompletionChunk {
       /**
        * The role of the delta.
        */
-      role?: string | null;
+      role?: 'developer' | 'system' | 'user' | 'assistant' | 'tool' | null;
 
       /**
        * The tool calls of the delta.
@@ -124,45 +133,39 @@ export namespace ChatCompletionChunk {
         /**
          * Unique identifier for the tool call.
          */
-        id?: string | null;
+        id: string;
 
         /**
-         * Function call details for OpenAI-compatible tool calls.
+         * Function call details.
          */
-        function?: ToolCall.Function | null;
-
-        /**
-         * Index of the tool call in the list.
-         */
-        index?: number | null;
+        function: ToolCall.Function;
 
         /**
          * Must be 'function' to identify this as a function call.
          */
-        type?: 'function';
+        type: 'function';
       }
 
       export namespace ToolCall {
         /**
-         * Function call details for OpenAI-compatible tool calls.
+         * Function call details.
          */
         export interface Function {
           /**
            * Arguments to pass to the function as a JSON string.
            */
-          arguments?: string | null;
+          arguments: string;
 
           /**
            * Name of the function to call.
            */
-          name?: string | null;
+          name: string;
         }
       }
     }
 
     /**
-     * The log probabilities for the tokens in the message from an OpenAI-compatible
-     * chat completion response.
+     * The log probabilities for the tokens in the message.
      */
     export interface Logprobs {
       /**
@@ -278,54 +281,54 @@ export namespace ChatCompletionChunk {
   }
 
   /**
-   * Usage information for OpenAI chat completion.
+   * Token usage information (typically included in final chunk with stream_options).
    */
   export interface Usage {
     /**
      * Number of tokens in the completion.
      */
-    completion_tokens: number;
+    completion_tokens?: number;
 
     /**
-     * Number of tokens in the prompt.
+     * Detailed breakdown of output token usage.
      */
-    prompt_tokens: number;
+    completion_tokens_details?: Usage.CompletionTokensDetails;
 
     /**
-     * Total tokens used (prompt + completion).
+     * Number of tokens in the prompt.
      */
-    total_tokens: number;
+    prompt_tokens?: number;
 
     /**
-     * Token details for output tokens in OpenAI chat completion usage.
+     * Detailed breakdown of input token usage.
      */
-    completion_tokens_details?: Usage.CompletionTokensDetails | null;
+    prompt_tokens_details?: Usage.PromptTokensDetails;
 
     /**
-     * Token details for prompt tokens in OpenAI chat completion usage.
+     * Total tokens used (prompt + completion).
      */
-    prompt_tokens_details?: Usage.PromptTokensDetails | null;
+    total_tokens?: number;
   }
 
   export namespace Usage {
     /**
-     * Token details for output tokens in OpenAI chat completion usage.
+     * Detailed breakdown of output token usage.
      */
     export interface CompletionTokensDetails {
       /**
        * Number of tokens used for reasoning (o1/o3 models).
        */
-      reasoning_tokens?: number | null;
+      reasoning_tokens?: number;
     }
 
     /**
-     * Token details for prompt tokens in OpenAI chat completion usage.
+     * Detailed breakdown of input token usage.
      */
     export interface PromptTokensDetails {
       /**
        * Number of tokens retrieved from cache.
        */
-      cached_tokens?: number | null;
+      cached_tokens?: number;
     }
   }
 }