npm - @google/genai - Versions diffs - 0.4.0 → 0.6.0 - Mend

@google/genai 0.4.0 → 0.6.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (13) hide show

package/README.md CHANGED Viewed

@@ -8,24 +8,25 @@
 ----------------------
-The Google Gen AI JavaScript SDK is an **experimental SDK** designed for
+The Google Gen AI JavaScript SDK is designed for
 TypeScript and JavaScript developers to build applications powered by Gemini. The SDK
 supports both the [Gemini Developer API](https://ai.google.dev/gemini-api/docs)
 and [Vertex AI](https://cloud.google.com/vertex-ai/generative-ai/docs/learn/overview).
 The Google Gen AI SDK is designed to work with Gemini 2.0 features.
-> [!CAUTION] **Experimental SDK:** This SDK is under active development and may
-  experience breaking changes.
+> [!NOTE]
+> **SDK Preview:**
+> See: [Preview Launch](#preview-launch).
-> [!CAUTION] **API Key Security:** Avoid exposing API keys in client-side code.
-  Use server-side implementations in production environments.
+> [!CAUTION]
+> **API Key Security:** Avoid exposing API keys in client-side code.
+> Use server-side implementations in production environments.
 ## Prerequisites
 * Node.js version 18 or later
-* `pnpm` or `npm`
 ## Installation
@@ -62,30 +63,31 @@ main();
 The package contents are also available unzipped in the
 `package/` directory of the bucket, so an equivalent web example is:
-```
+```html
 <!DOCTYPE html>
 <html lang="en">
-<head>
-    <meta charset="UTF-8">
-    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
     <title>Using My Package</title>
-</head>
-<body>
+  </head>
+  <body>
     <script type="module">
-        import {GoogleGenAI, Type} from 'dist/web/index.mjs';
-        const ai = new GoogleGenAI({apiKey:"YOUR_API_KEY"});
-        async function main() {
-          const response = await ai.models.generateContent({
-            model: 'gemini-2.0-flash-001',
-            contents: 'Why is the sky blue?',
-          });
-          console.log(response.text());
-        }
+      import {GoogleGenAI} from 'https://cdn.jsdelivr.net/npm/@google/genai@latest/+esm'
+          const ai = new GoogleGenAI({apiKey:"GEMINI_API_KEY"});
-        main();
+          async function main() {
+            const response = await ai.models.generateContent({
+              model: 'gemini-2.0-flash-001',
+              contents: 'Why is the sky blue?',
+            });
+            console.log(response.text);
+          }
+          main();
     </script>
-</body>
+  </body>
 </html>
 ```
@@ -108,8 +110,9 @@ const ai = new GoogleGenAI({apiKey: 'GEMINI_API_KEY'});
 #### Browser
-> [!CAUTION] **API Key Security:** Avoid exposing API keys in client-side code.
-   Use server-side implementations in production environments.
+> [!CAUTION]
+> **API Key Security:** Avoid exposing API keys in client-side code.
+>   Use server-side implementations in production environments.
 In the browser the initialization code is identical:
@@ -138,26 +141,26 @@ const ai = new GoogleGenAI({
 All API features are accessed through an instance of the `GoogleGenAI` classes.
 The submodules bundle together related API methods:
-- [`client.models`](https://googleapis.github.io/js-genai/classes/models.Models.html):
+- [`ai.models`](https://googleapis.github.io/js-genai/classes/models.Models.html):
   Use `models` to query models (`generateContent`, `generateImages`, ...), or
   examine their metadata.
-- [`client.caches`](https://googleapis.github.io/js-genai/classes/caches.Caches.html):
+- [`ai.caches`](https://googleapis.github.io/js-genai/classes/caches.Caches.html):
   Create and manage `caches` to reduce costs when repeatedly using the same
   large prompt prefix.
-- [`client.chats`](https://googleapis.github.io/js-genai/classes/chats.Chats.html):
+- [`ai.chats`](https://googleapis.github.io/js-genai/classes/chats.Chats.html):
   Create local stateful `chat` objects to simplify multi turn interactions.
-- [`client.files`](https://googleapis.github.io/js-genai/classes/files.Files.html):
+- [`ai.files`](https://googleapis.github.io/js-genai/classes/files.Files.html):
   Upload `files` to the API and reference them in your prompts.
-  This reduces bandwidth if you use a file many times, and handles files too
+  This reduces bandwidth if you use a file many times, and handles files too
   large to fit inline with your prompt.
-- [`client.live`](https://googleapis.github.io/js-genai/classes/live.Live.html):
+- [`ai.live`](https://googleapis.github.io/js-genai/classes/live.Live.html):
   Start a `live` session for real time interaction, allows text + audio + video
   input, and text or audio output.
 ## Samples
 More samples can be found in the
-[github samples directory](https://github.com/google-gemini/generative-ai-js/tree/main/sdk-samples).
+[github samples directory](https://github.com/googleapis/js-genai/tree/main/sdk-samples).
 ### Streaming
@@ -242,3 +245,11 @@ async function main() {
 main();
 ```
+## Preview Launch
+The SDK is curently in a preview launch stage, per [Google's launch stages](https://cloud.google.com/products?hl=en#section-22) this means:
+> At Preview, products or features are ready for testing by customers. Preview offerings are often publicly announced, but are not necessarily feature-complete, and no SLAs or technical support commitments are provided for these. Unless stated otherwise by Google, Preview offerings are intended for use in test environments only. The average Preview stage lasts about six months.

package/dist/genai.d.ts CHANGED Viewed

@@ -261,10 +261,12 @@ export declare interface Candidate {
     /** Number of tokens for this candidate.
      */
     tokenCount?: number;
+    /** The reason why the model stopped generating tokens.
+     If empty, the model has not stopped generating the tokens.
+     */
+    finishReason?: FinishReason;
     /** Output only. Average log probability score of the candidate. */
     avgLogprobs?: number;
-    /** Output only. The reason why the model stopped generating tokens. If empty, the model has not stopped generating the tokens. */
-    finishReason?: FinishReason;
     /** Output only. Metadata specifies sources used to ground generated content. */
     groundingMetadata?: GroundingMetadata;
     /** Output only. Index of the candidate. */
@@ -375,7 +377,7 @@ export declare class Chats {
      * @remarks
      * The config in the params will be used for all requests within the chat
      * session unless overridden by a per-request `config` in
-     * {@link ./types.SendMessageParameters}.
+     * @see {@link types.SendMessageParameters#config}.
      *
      * @param params - Parameters for creating a chat session.
      * @returns A new chat session.
@@ -496,12 +498,6 @@ export declare interface ContentEmbeddingStatistics {
 export declare type ContentListUnion = ContentUnion[] | ContentUnion;
-/** @internal */
-export declare function contentToMldev(apiClient: ApiClient, fromObject: types.Content): Record<string, unknown>;
-/** @internal */
-export declare function contentToVertex(apiClient: ApiClient, fromObject: types.Content): Record<string, unknown>;
 export declare type ContentUnion = Content | PartUnion[] | PartUnion;
 /** Configuration for a Control reference image. */
@@ -989,7 +985,8 @@ export declare enum FinishReason {
     BLOCKLIST = "BLOCKLIST",
     PROHIBITED_CONTENT = "PROHIBITED_CONTENT",
     SPII = "SPII",
-    MALFORMED_FUNCTION_CALL = "MALFORMED_FUNCTION_CALL"
+    MALFORMED_FUNCTION_CALL = "MALFORMED_FUNCTION_CALL",
+    IMAGE_SAFETY = "IMAGE_SAFETY"
 }
 declare function formatMap(templateString: string, valueMap: Record<string, unknown>): string;
@@ -1317,13 +1314,25 @@ export declare class GenerateContentResponsePromptFeedback {
 /** Usage metadata about response(s). */
 export declare class GenerateContentResponseUsageMetadata {
+    /** Output only. List of modalities of the cached content in the request input. */
+    cacheTokensDetails?: ModalityTokenCount[];
     /** Output only. Number of tokens in the cached part in the input (the cached content). */
     cachedContentTokenCount?: number;
     /** Number of tokens in the response(s). */
     candidatesTokenCount?: number;
+    /** Output only. List of modalities that were returned in the response. */
+    candidatesTokensDetails?: ModalityTokenCount[];
     /** Number of tokens in the request. When `cached_content` is set, this is still the total effective prompt size meaning this includes the number of tokens in the cached content. */
     promptTokenCount?: number;
-    /** Total token count for prompt and response candidates. */
+    /** Output only. List of modalities that were processed in the request input. */
+    promptTokensDetails?: ModalityTokenCount[];
+    /** Output only. Number of tokens present in thoughts output. */
+    thoughtsTokenCount?: number;
+    /** Output only. Number of tokens present in tool-use prompt(s). */
+    toolUsePromptTokenCount?: number;
+    /** Output only. List of modalities that were processed for tool-use request inputs. */
+    toolUsePromptTokensDetails?: ModalityTokenCount[];
+    /** Total token count for prompt, response candidates, and tool-use prompts (if present). */
     totalTokenCount?: number;
 }
@@ -1883,6 +1892,11 @@ export declare class Live {
      @experimental
+     @remarks
+     If using the Gemini API, Live is currently only supported behind API
+     version `v1alpha`. Ensure that the API version is set to `v1alpha` when
+     initializing the SDK if relying on the Gemini API.
      @param params - The parameters for establishing a connection to the model.
      @return A live session.
@@ -2055,6 +2069,28 @@ export declare interface LiveConnectParameters {
     config?: LiveConnectConfig;
 }
+/** Parameters for sending client content to the live API. */
+export declare interface LiveSendClientContentParameters {
+    /** Client content to send to the session. */
+    turns?: ContentListUnion;
+    /** If true, indicates that the server content generation should start with
+     the currently accumulated prompt. Otherwise, the server will await
+     additional messages before starting generation. */
+    turnComplete?: boolean;
+}
+/** Parameters for sending realtime input to the live API. */
+export declare interface LiveSendRealtimeInputParameters {
+    /** Realtime input to send to the session. */
+    media: Blob_2;
+}
+/** Parameters for sending tool responses to the live API. */
+export declare class LiveSendToolResponseParameters {
+    /** Tool responses to send to the session. */
+    functionResponses: FunctionResponse[] | FunctionResponse;
+}
 /** Incremental server update generated by the model in response to client messages.
  Content is generated as quickly as possible, and not in real time. Clients
@@ -2182,6 +2218,14 @@ export declare enum Modality {
     AUDIO = "AUDIO"
 }
+/** Represents token counting info for a single modality. */
+export declare interface ModalityTokenCount {
+    /** The modality associated with this token count. */
+    modality?: Modality;
+    /** Number of tokens. */
+    tokenCount?: number;
+}
 export declare enum Mode {
     MODE_UNSPECIFIED = "MODE_UNSPECIFIED",
     MODE_DYNAMIC = "MODE_DYNAMIC"
@@ -2371,7 +2415,7 @@ export declare enum Outcome {
  * SPDX-License-Identifier: Apache-2.0
  */
 /**
- * @fileoverview Pagers for the GenAI List APIs.
+ * Pagers for the GenAI List APIs.
  */
 declare enum PagedItem {
     PAGED_ITEM_BATCH_JOBS = "batchJobs",
@@ -2752,12 +2796,13 @@ export declare class Session {
      `turnComplete`.
      - `turns` will be converted to a `Content[]`
-     - `turnComplete: true` indicates that you are done sending content and
-     expect a response.
+     - `turnComplete: true` [default] indicates that you are done sending
+     content and expect a response. If `turnComplete: false`, the server
+     will wait for additional messages before starting generation.
      @experimental
-     @remanks
+     @remarks
      There are two ways to send messages to the live API:
      `sendClientContent` and `sendRealtimeInput`.
@@ -2793,7 +2838,7 @@ export declare class Session {
      ```
      @experimental
      */
-    sendClientContent(params: types.SessionSendClientContentParameters): void;
+    sendClientContent(params: types.LiveSendClientContentParameters): void;
     /**
      Send a realtime message over the established connection.
@@ -2803,7 +2848,7 @@ export declare class Session {
      @experimental
-     @remanks
+     @remarks
      Use `sendRealtimeInput` for realtime audio chunks and video frames (images).
      With `sendRealtimeInput` the api will respond to audio automatically
@@ -2816,7 +2861,7 @@ export declare class Session {
      Note: The Call signature expects a `Blob` object, but only a subset
      of audio and image mimetypes are allowed.
      */
-    sendRealtimeInput(params: types.SessionSendRealtimeInputParameters): void;
+    sendRealtimeInput(params: types.LiveSendRealtimeInputParameters): void;
     /**
      Send a function response message over the established connection.
@@ -2824,14 +2869,14 @@ export declare class Session {
      - `functionResponses` will be converted to a `functionResponses[]`
-     @remanks
+     @remarks
      Use `sendFunctionResponse` to reply to `LiveServerToolCall` from the server.
-     Use {@link LiveConnectConfig#tools} to configure the callable functions.
+     Use {@link types.LiveConnectConfig#tools} to configure the callable functions.
      @experimental
      */
-    sendToolResponse(params: types.SessionSendToolResponseParameters): void;
+    sendToolResponse(params: types.LiveSendToolResponseParameters): void;
     /**
      Terminates the WebSocket connection.
@@ -2852,28 +2897,6 @@ export declare class Session {
     close(): void;
 }
-/** Parameters for sending client content to the live API. */
-export declare interface SessionSendClientContentParameters {
-    /** Client content to send to the session. */
-    turns?: ContentListUnion;
-    /** If true, indicates that the server content generation should start with
-     the currently accumulated prompt. Otherwise, the server will await
-     additional messages before starting generation. */
-    turnComplete?: boolean;
-}
-/** Parameters for sending realtime input to the live API. */
-export declare interface SessionSendRealtimeInputParameters {
-    /** Realtime input to send to the session. */
-    media: Blob_2;
-}
-/** Parameters for sending tool responses to the live API. */
-export declare class SessionSendToolResponseParameters {
-    /** Tool responses to send to the session. */
-    functionResponses: FunctionResponse | FunctionResponse[];
-}
 declare function setValueByPath(data: Record<string, unknown>, keys: string[], value: unknown): void;
 /** The speech generation configuration. */
@@ -3021,12 +3044,6 @@ export declare interface ToolConfig {
 export declare type ToolListUnion = Tool[];
-/** @internal */
-export declare function toolToMldev(apiClient: ApiClient, fromObject: types.Tool): Record<string, unknown>;
-/** @internal */
-export declare function toolToVertex(apiClient: ApiClient, fromObject: types.Tool): Record<string, unknown>;
 export declare enum Type {
     TYPE_UNSPECIFIED = "TYPE_UNSPECIFIED",
     STRING = "STRING",
@@ -3059,6 +3076,7 @@ declare namespace types {
         HarmProbability,
         HarmSeverity,
         BlockedReason,
+        Modality,
         State,
         DynamicRetrievalConfigMode,
         FunctionCallingConfigMode,
@@ -3071,7 +3089,6 @@ declare namespace types {
         MaskReferenceMode,
         ControlReferenceType,
         SubjectReferenceType,
-        Modality,
         VideoMetadata,
         CodeExecutionResult,
         ExecutableCode,
@@ -3122,6 +3139,7 @@ declare namespace types {
         SafetyRating,
         Candidate,
         GenerateContentResponsePromptFeedback,
+        ModalityTokenCount,
         GenerateContentResponseUsageMetadata,
         GenerateContentResponse,
         EmbedContentConfig,
@@ -3203,9 +3221,9 @@ declare namespace types {
         LiveConnectParameters,
         CreateChatParameters,
         SendMessageParameters,
-        SessionSendClientContentParameters,
-        SessionSendRealtimeInputParameters,
-        SessionSendToolResponseParameters,
+        LiveSendClientContentParameters,
+        LiveSendRealtimeInputParameters,
+        LiveSendToolResponseParameters,
         PartUnion,
         PartListUnion,
         ContentUnion,