feat: Support thinking settings for Gemini models (#19591)

packages/@n8n/nodes-langchain/nodes/llms/LmChatGoogleGemini/LmChatGoogleGemini.node.ts

@@ -11,7 +11,7 @@ import type {
 
 import { getConnectionHintNoticeField } from '@utils/sharedFields';
 
-import { additionalOptions } from '../gemini-common/additional-options';
+import { getAdditionalOptions } from '../gemini-common/additional-options';
 import { makeN8nLlmFailedAttemptHandler } from '../n8nLlmFailedAttemptHandler';
 import { N8nLlmTracing } from '../N8nLlmTracing';
 
@@ -119,7 +119,9 @@ export class LmChatGoogleGemini implements INodeType {
 				},
 				default: 'models/gemini-2.5-flash',
 			},
-			additionalOptions,
+			// thinking budget not supported in @langchain/google-genai
+			// as it utilises the old google generative ai SDK
+			getAdditionalOptions({ supportsThinkingBudget: false }),
 		],
 	};
 

packages/@n8n/nodes-langchain/nodes/llms/LmChatGoogleVertex/LmChatGoogleVertex.node.ts

@@ -1,6 +1,6 @@
-import type { SafetySetting } from '@google/generative-ai';
 import { ProjectsClient } from '@google-cloud/resource-manager';
-import { ChatVertexAI } from '@langchain/google-vertexai';
+import type { GoogleAISafetySetting } from '@langchain/google-common';
+import { ChatVertexAI, type ChatVertexAIInput } from '@langchain/google-vertexai';
 import { formatPrivateKey } from 'n8n-nodes-base/dist/utils/utilities';
 import {
 	NodeConnectionTypes,
@@ -11,12 +11,13 @@ import {
 	type ILoadOptionsFunctions,
 	type JsonObject,
 	NodeOperationError,
+	validateNodeParameters,
 } from 'n8n-workflow';
 
 import { getConnectionHintNoticeField } from '@utils/sharedFields';
 
 import { makeErrorFromStatus } from './error-handling';
-import { additionalOptions } from '../gemini-common/additional-options';
+import { getAdditionalOptions } from '../gemini-common/additional-options';
 import { makeN8nLlmFailedAttemptHandler } from '../n8nLlmFailedAttemptHandler';
 import { N8nLlmTracing } from '../N8nLlmTracing';
 
@@ -90,7 +91,7 @@ export class LmChatGoogleVertex implements INodeType {
 					'The model which will generate the completion. <a href="https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models">Learn more</a>.',
 				default: 'gemini-2.5-flash',
 			},
-			additionalOptions,
+			getAdditionalOptions({ supportsThinkingBudget: true }),
 		],
 	};
 
@@ -143,21 +144,29 @@ export class LmChatGoogleVertex implements INodeType {
 			temperature: 0.4,
 			topK: 40,
 			topP: 0.9,
-		}) as {
-			maxOutputTokens: number;
-			temperature: number;
-			topK: number;
-			topP: number;
-		};
+		});
+
+		// Validate options parameter
+		validateNodeParameters(
+			options,
+			{
+				maxOutputTokens: { type: 'number', required: false },
+				temperature: { type: 'number', required: false },
+				topK: { type: 'number', required: false },
+				topP: { type: 'number', required: false },
+				thinkingBudget: { type: 'number', required: false },
+			},
+			this.getNode(),
+		);
 
 		const safetySettings = this.getNodeParameter(
 			'options.safetySettings.values',
 			itemIndex,
 			null,
-		) as SafetySetting[];
+		) as GoogleAISafetySetting[];
 
 		try {
-			const model = new ChatVertexAI({
+			const modelConfig: ChatVertexAIInput = {
 				authOptions: {
 					projectId,
 					credentials: {
@@ -186,7 +195,14 @@ export class LmChatGoogleVertex implements INodeType {
 
 					throw error;
 				}),
-			});
+			};
+
+			// Add thinkingBudget if specified
+			if (options.thinkingBudget !== undefined) {
+				modelConfig.thinkingBudget = options.thinkingBudget;
+			}
+
+			const model = new ChatVertexAI(modelConfig);
 
 			return {
 				response: model,

packages/@n8n/nodes-langchain/nodes/llms/LmChatGoogleVertex/test/LmChatGoogleVertex.test.ts

@@ -0,0 +1,149 @@
+import { ChatVertexAI } from '@langchain/google-vertexai';
+import { createMockExecuteFunction } from 'n8n-nodes-base/test/nodes/Helpers';
+import type { INode, ISupplyDataFunctions } from 'n8n-workflow';
+
+import { makeN8nLlmFailedAttemptHandler } from '../../n8nLlmFailedAttemptHandler';
+import { N8nLlmTracing } from '../../N8nLlmTracing';
+import { LmChatGoogleVertex } from '../LmChatGoogleVertex.node';
+
+jest.mock('@langchain/google-vertexai');
+jest.mock('../../N8nLlmTracing');
+jest.mock('../../n8nLlmFailedAttemptHandler');
+jest.mock('n8n-nodes-base/dist/utils/utilities', () => ({
+	formatPrivateKey: jest.fn().mockImplementation((key: string) => key),
+}));
+
+const MockedChatVertexAI = jest.mocked(ChatVertexAI);
+const MockedN8nLlmTracing = jest.mocked(N8nLlmTracing);
+const mockedMakeN8nLlmFailedAttemptHandler = jest.mocked(makeN8nLlmFailedAttemptHandler);
+
+describe('LmChatGoogleVertex - Thinking Budget', () => {
+	let lmChatGoogleVertex: LmChatGoogleVertex;
+	let mockContext: jest.Mocked<ISupplyDataFunctions>;
+
+	const mockNode: INode = {
+		id: '1',
+		name: 'Google Vertex Chat Model',
+		typeVersion: 1,
+		type: 'n8n-nodes-langchain.lmChatGoogleVertex',
+		position: [0, 0],
+		parameters: {},
+	};
+
+	const setupMockContext = () => {
+		mockContext = createMockExecuteFunction<ISupplyDataFunctions>(
+			{},
+			mockNode,
+		) as jest.Mocked<ISupplyDataFunctions>;
+
+		mockContext.getCredentials = jest.fn().mockResolvedValue({
+			privateKey: 'test-private-key',
+			email: 'test@n8n.io',
+			region: 'us-central1',
+		});
+		mockContext.getNode = jest.fn().mockReturnValue(mockNode);
+		mockContext.getNodeParameter = jest.fn();
+
+		MockedN8nLlmTracing.mockImplementation(() => ({}) as unknown as N8nLlmTracing);
+		mockedMakeN8nLlmFailedAttemptHandler.mockReturnValue(jest.fn());
+
+		return mockContext;
+	};
+
+	beforeEach(() => {
+		lmChatGoogleVertex = new LmChatGoogleVertex();
+		jest.clearAllMocks();
+	});
+
+	afterEach(() => {
+		jest.clearAllMocks();
+	});
+
+	describe('supplyData - thinking budget parameter passing', () => {
+		it('should not include thinkingBudget in model config when not specified', async () => {
+			const mockContext = setupMockContext();
+
+			mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
+				if (paramName === 'modelName') return 'gemini-2.5-flash';
+				if (paramName === 'projectId') return 'test-project';
+				if (paramName === 'options') {
+					// Return options without thinkingBudget
+					return {
+						maxOutputTokens: 2048,
+						temperature: 0.4,
+						topK: 40,
+						topP: 0.9,
+					};
+				}
+				if (paramName === 'options.safetySettings.values') return null;
+				return undefined;
+			});
+
+			await lmChatGoogleVertex.supplyData.call(mockContext, 0);
+			expect(MockedChatVertexAI).toHaveBeenCalledTimes(1);
+			const callArgs = MockedChatVertexAI.mock.calls[0][0];
+			expect(callArgs).not.toHaveProperty('thinkingBudget');
+			expect(callArgs).toMatchObject({
+				authOptions: {
+					projectId: 'test-project',
+					credentials: {
+						client_email: 'test@n8n.io',
+						private_key: 'test-private-key',
+					},
+				},
+				location: 'us-central1',
+				model: 'gemini-2.5-flash',
+				topK: 40,
+				topP: 0.9,
+				temperature: 0.4,
+				maxOutputTokens: 2048,
+			});
+		});
+
+		it('should include thinkingBudget in model config when specified', async () => {
+			const mockContext = setupMockContext();
+			const expectedThinkingBudget = 1024;
+
+			mockContext.getNodeParameter = jest.fn().mockImplementation((paramName: string) => {
+				if (paramName === 'modelName') return 'gemini-2.5-flash';
+				if (paramName === 'projectId') return 'test-project';
+				if (paramName === 'options') {
+					// Return options with thinkingBudget
+					return {
+						maxOutputTokens: 2048,
+						temperature: 0.4,
+						topK: 40,
+						topP: 0.9,
+						thinkingBudget: expectedThinkingBudget,
+					};
+				}
+				if (paramName === 'options.safetySettings.values') return null;
+				return undefined;
+			});
+
+			await lmChatGoogleVertex.supplyData.call(mockContext, 0);
+			expect(MockedChatVertexAI).toHaveBeenCalledWith(
+				expect.objectContaining({
+					authOptions: {
+						projectId: 'test-project',
+						credentials: {
+							client_email: 'test@n8n.io',
+							private_key: 'test-private-key',
+						},
+					},
+					location: 'us-central1',
+					model: 'gemini-2.5-flash',
+					topK: 40,
+					topP: 0.9,
+					temperature: 0.4,
+					maxOutputTokens: 2048,
+					thinkingBudget: expectedThinkingBudget,
+					// eslint-disable-next-line @typescript-eslint/no-unsafe-assignment
+					callbacks: expect.arrayContaining([expect.any(Object)]),
+					// eslint-disable-next-line @typescript-eslint/no-unsafe-assignment
+					onFailedAttempt: expect.any(Function),
+				}),
+			);
+		});
+	});
+});

packages/@n8n/nodes-langchain/nodes/llms/gemini-common/additional-options.ts

@@ -1,88 +1,107 @@
-import type { HarmBlockThreshold, HarmCategory } from '@google/generative-ai';
+import type { HarmBlockThreshold, HarmCategory } from '@google/genai';
 import type { INodeProperties } from 'n8n-workflow';
 
 import { harmCategories, harmThresholds } from './safety-options';
 
-export const additionalOptions: INodeProperties = {
-	displayName: 'Options',
-	name: 'options',
-	placeholder: 'Add Option',
-	description: 'Additional options to add',
-	type: 'collection',
-	default: {},
-	options: [
-		{
-			displayName: 'Maximum Number of Tokens',
-			name: 'maxOutputTokens',
-			default: 2048,
-			description: 'The maximum number of tokens to generate in the completion',
-			type: 'number',
-		},
-		{
-			displayName: 'Sampling Temperature',
-			name: 'temperature',
-			default: 0.4,
-			typeOptions: { maxValue: 1, minValue: 0, numberPrecision: 1 },
-			description:
-				'Controls randomness: Lowering results in less random completions. As the temperature approaches zero, the model will become deterministic and repetitive.',
-			type: 'number',
-		},
-		{
-			displayName: 'Top K',
-			name: 'topK',
-			default: 32,
-			typeOptions: { maxValue: 40, minValue: -1, numberPrecision: 1 },
-			description:
-				'Used to remove "long tail" low probability responses. Defaults to -1, which disables it.',
-			type: 'number',
-		},
-		{
-			displayName: 'Top P',
-			name: 'topP',
-			default: 1,
-			typeOptions: { maxValue: 1, minValue: 0, numberPrecision: 1 },
-			description:
-				'Controls diversity via nucleus sampling: 0.5 means half of all likelihood-weighted options are considered. We generally recommend altering this or temperature but not both.',
-			type: 'number',
-		},
-
-		// Safety Settings
-		{
-			displayName: 'Safety Settings',
-			name: 'safetySettings',
-			type: 'fixedCollection',
-			typeOptions: { multipleValues: true },
-			default: {
-				values: {
-					category: harmCategories[0].name as HarmCategory,
-					threshold: harmThresholds[0].name as HarmBlockThreshold,
-				},
+export function getAdditionalOptions({
+	supportsThinkingBudget,
+}: { supportsThinkingBudget: boolean }) {
+	const baseOptions: INodeProperties = {
+		displayName: 'Options',
+		name: 'options',
+		placeholder: 'Add Option',
+		description: 'Additional options to add',
+		type: 'collection',
+		default: {},
+		options: [
+			{
+				displayName: 'Maximum Number of Tokens',
+				name: 'maxOutputTokens',
+				default: 2048,
+				description: 'The maximum number of tokens to generate in the completion',
+				type: 'number',
 			},
-			placeholder: 'Add Option',
-			options: [
-				{
-					name: 'values',
-					displayName: 'Values',
-					values: [
-						{
-							displayName: 'Safety Category',
-							name: 'category',
-							type: 'options',
-							description: 'The category of harmful content to block',
-							default: 'HARM_CATEGORY_UNSPECIFIED',
-							options: harmCategories,
-						},
-						{
-							displayName: 'Safety Threshold',
-							name: 'threshold',
-							type: 'options',
-							description: 'The threshold of harmful content to block',
-							default: 'HARM_BLOCK_THRESHOLD_UNSPECIFIED',
-							options: harmThresholds,
-						},
-					],
+			{
+				displayName: 'Sampling Temperature',
+				name: 'temperature',
+				default: 0.4,
+				typeOptions: { maxValue: 1, minValue: 0, numberPrecision: 1 },
+				description:
+					'Controls randomness: Lowering results in less random completions. As the temperature approaches zero, the model will become deterministic and repetitive.',
+				type: 'number',
+			},
+			{
+				displayName: 'Top K',
+				name: 'topK',
+				default: 32,
+				typeOptions: { maxValue: 40, minValue: -1, numberPrecision: 1 },
+				description:
+					'Used to remove "long tail" low probability responses. Defaults to -1, which disables it.',
+				type: 'number',
+			},
+			{
+				displayName: 'Top P',
+				name: 'topP',
+				default: 1,
+				typeOptions: { maxValue: 1, minValue: 0, numberPrecision: 1 },
+				description:
+					'Controls diversity via nucleus sampling: 0.5 means half of all likelihood-weighted options are considered. We generally recommend altering this or temperature but not both.',
+				type: 'number',
+			},
+			// Safety Settings
+			{
+				displayName: 'Safety Settings',
+				name: 'safetySettings',
+				type: 'fixedCollection',
+				typeOptions: { multipleValues: true },
+				default: {
+					values: {
+						category: harmCategories[0].name as HarmCategory,
+						threshold: harmThresholds[0].name as HarmBlockThreshold,
+					},
 				},
-			],
-		},
-	],
-};
+				placeholder: 'Add Option',
+				options: [
+					{
+						name: 'values',
+						displayName: 'Values',
+						values: [
+							{
+								displayName: 'Safety Category',
+								name: 'category',
+								type: 'options',
+								description: 'The category of harmful content to block',
+								default: 'HARM_CATEGORY_UNSPECIFIED',
+								options: harmCategories,
+							},
+							{
+								displayName: 'Safety Threshold',
+								name: 'threshold',
+								type: 'options',
+								description: 'The threshold of harmful content to block',
+								default: 'HARM_BLOCK_THRESHOLD_UNSPECIFIED',
+								options: harmThresholds,
+							},
+						],
+					},
+				],
+			},
+		],
+	};
+	// only supported in the new google genai SDK
+	if (supportsThinkingBudget) {
+		baseOptions.options?.push({
+			displayName: 'Thinking Budget',
+			name: 'thinkingBudget',
+			default: undefined,
+			description:
+				'Controls reasoning tokens for thinking models. Set to 0 to disable automatic thinking. Set to -1 for dynamic thinking. Leave empty for auto mode.',
+			type: 'number',
+			typeOptions: {
+				minValue: -1,
+				numberPrecision: 0,
+			},
+		});
+	}
+	return baseOptions;
+}

packages/@n8n/nodes-langchain/nodes/vendors/GoogleGemini/GoogleGemini.node.test.ts

@@ -80,7 +80,7 @@ describe('GoogleGemini Node', () => {
 			expect(apiRequestMock).toHaveBeenCalledWith(
 				'POST',
 				'/v1beta/models/gemini-2.5-flash:generateContent',
-				{
+				expect.objectContaining({
 					body: {
 						contents: [
 							{
@@ -107,6 +107,76 @@ describe('GoogleGemini Node', () => {
 							parts: [{ text: 'You are a helpful assistant.' }],
 						},
 					},
+				}),
+			);
+		});
+
+		it('should include thinking options when the thinking budget is specified', async () => {
+			executeFunctionsMock.getNodeParameter.mockImplementation((parameter: string) => {
+				switch (parameter) {
+					case 'modelId':
+						return 'models/gemini-2.5-flash';
+					case 'messages.values':
+						return [{ role: 'user', content: 'Hello, world!' }];
+					case 'simplify':
+						return true;
+					case 'jsonOutput':
+						return false;
+					case 'options':
+						return {
+							thinkingBudget: 1024,
+							maxOutputTokens: 100,
+							temperature: 0.5,
+						};
+					default:
+						return undefined;
+				}
+			});
+			executeFunctionsMock.getNodeInputs.mockReturnValue([{ type: 'main' }]);
+			apiRequestMock.mockResolvedValue({
+				candidates: [
+					{
+						content: {
+							parts: [{ text: 'Hello with thinking!' }],
+							role: 'model',
+						},
+					},
+				],
+			});
+
+			const result = await text.message.execute.call(executeFunctionsMock, 0);
+
+			expect(result).toEqual([
+				{
+					json: {
+						content: {
+							parts: [{ text: 'Hello with thinking!' }],
+							role: 'model',
+						},
+					},
+					pairedItem: { item: 0 },
+				},
+			]);
+			expect(apiRequestMock).toHaveBeenCalledWith(
+				'POST',
+				'/v1beta/models/gemini-2.5-flash:generateContent',
+				{
+					body: {
+						contents: [
+							{
+								parts: [{ text: 'Hello, world!' }],
+								role: 'user',
+							},
+						],
+						tools: [],
+						generationConfig: {
+							maxOutputTokens: 100,
+							temperature: 0.5,
+							thinkingConfig: {
+								thinkingBudget: 1024,
+							},
+						},
+					},
 				},
 			);
 		});

packages/@n8n/nodes-langchain/nodes/vendors/GoogleGemini/actions/audio/transcribe.operation.ts

@@ -1,7 +1,11 @@
 import type { IExecuteFunctions, INodeExecutionData, INodeProperties } from 'n8n-workflow';
 import { updateDisplayOptions } from 'n8n-workflow';
 
-import type { Content, GenerateContentResponse } from '../../helpers/interfaces';
+import type {
+	Content,
+	GenerateContentRequest,
+	GenerateContentResponse,
+} from '../../helpers/interfaces';
 import { downloadFile, uploadFile } from '../../helpers/utils';
 import { apiRequest } from '../../transport';
 import { modelRLC } from '../descriptions';
@@ -157,7 +161,7 @@ export async function execute(this: IExecuteFunctions, i: number): Promise<INode
 	}${options.endTime ? ` to ${options.endTime as string}` : ''}`;
 	contents[0].parts.push({ text });
 
-	const body = {
+	const body: GenerateContentRequest = {
 		contents,
 	};
 

packages/@n8n/nodes-langchain/nodes/vendors/GoogleGemini/actions/image/generate.operation.ts

@@ -1,7 +1,12 @@
 import type { IExecuteFunctions, INodeExecutionData, INodeProperties } from 'n8n-workflow';
 import { NodeOperationError, updateDisplayOptions } from 'n8n-workflow';
 
-import type { GenerateContentResponse, ImagenResponse } from '../../helpers/interfaces';
+import {
+	type GenerateContentRequest,
+	type GenerateContentResponse,
+	type ImagenResponse,
+	Modality,
+} from '../../helpers/interfaces';
 import { apiRequest } from '../../transport';
 import { modelRLC } from '../descriptions';
 
@@ -67,9 +72,9 @@ export async function execute(this: IExecuteFunctions, i: number): Promise<INode
 
 	if (model.includes('gemini')) {
 		const generationConfig = {
-			responseModalities: ['IMAGE', 'TEXT'],
+			responseModalities: [Modality.IMAGE, Modality.TEXT],
 		};
-		const body = {
+		const body: GenerateContentRequest = {
 			contents: [
 				{
 					role: 'user',

packages/@n8n/nodes-langchain/nodes/vendors/GoogleGemini/actions/text/message.operation.ts

@@ -1,15 +1,22 @@
-import type {
-	IDataObject,
-	IExecuteFunctions,
-	INodeExecutionData,
-	INodeProperties,
+import {
+	type IDataObject,
+	type IExecuteFunctions,
+	type INodeExecutionData,
+	type INodeProperties,
+	validateNodeParameters,
 } from 'n8n-workflow';
 import { updateDisplayOptions } from 'n8n-workflow';
 import zodToJsonSchema from 'zod-to-json-schema';
 
 import { getConnectedTools } from '@utils/helpers';
 
-import type { GenerateContentResponse, Content, Tool } from '../../helpers/interfaces';
+import type {
+	GenerateContentRequest,
+	GenerateContentResponse,
+	Content,
+	Tool,
+	GenerateContentGenerationConfig,
+} from '../../helpers/interfaces';
 import { apiRequest } from '../../transport';
 import { modelRLC } from '../descriptions';
 
@@ -186,6 +193,18 @@ const properties: INodeProperties[] = [
 					numberPrecision: 0,
 				},
 			},
+			{
+				displayName: 'Thinking Budget',
+				name: 'thinkingBudget',
+				type: 'number',
+				default: undefined,
+				description:
+					'Controls reasoning tokens for thinking models. Set to 0 to disable automatic thinking. Set to -1 for dynamic thinking. Leave empty for auto mode.',
+				typeOptions: {
+					minValue: -1,
+					numberPrecision: 0,
+				},
+			},
 			{
 				displayName: 'Max Tool Calls Iterations',
 				name: 'maxToolsIterations',
@@ -224,8 +243,25 @@ export async function execute(this: IExecuteFunctions, i: number): Promise<INode
 	const simplify = this.getNodeParameter('simplify', i, true) as boolean;
 	const jsonOutput = this.getNodeParameter('jsonOutput', i, false) as boolean;
 	const options = this.getNodeParameter('options', i, {});
+	validateNodeParameters(
+		options,
+		{
+			systemMessage: { type: 'string', required: false },
+			codeExecution: { type: 'boolean', required: false },
+			frequencyPenalty: { type: 'number', required: false },
+			maxOutputTokens: { type: 'number', required: false },
+			candidateCount: { type: 'number', required: false },
+			presencePenalty: { type: 'number', required: false },
+			temperature: { type: 'number', required: false },
+			topP: { type: 'number', required: false },
+			topK: { type: 'number', required: false },
+			thinkingBudget: { type: 'number', required: false },
+			maxToolsIterations: { type: 'number', required: false },
+		},
+		this.getNode(),
+	);
 
-	const generationConfig = {
+	const generationConfig: GenerateContentGenerationConfig = {
 		frequencyPenalty: options.frequencyPenalty,
 		maxOutputTokens: options.maxOutputTokens,
 		candidateCount: options.candidateCount,
@@ -236,6 +272,13 @@ export async function execute(this: IExecuteFunctions, i: number): Promise<INode
 		responseMimeType: jsonOutput ? 'application/json' : undefined,
 	};
 
+	// Add thinkingConfig if thinkingBudget is specified
+	if (options.thinkingBudget !== undefined) {
+		generationConfig.thinkingConfig = {
+			thinkingBudget: options.thinkingBudget,
+		};
+	}
+
 	const nodeInputs = this.getNodeInputs();
 	const availableTools = nodeInputs.some((i) => i.type === 'ai_tool')
 		? await getConnectedTools(this, true)
@@ -267,7 +310,7 @@ export async function execute(this: IExecuteFunctions, i: number): Promise<INode
 		parts: [{ text: m.content }],
 		role: m.role,
 	}));
-	const body = {
+	const body: GenerateContentRequest = {
 		tools,
 		contents,
 		generationConfig,

packages/@n8n/nodes-langchain/nodes/vendors/GoogleGemini/helpers/baseAnalyze.ts

@@ -1,4 +1,8 @@
-import type { IExecuteFunctions, INodeExecutionData } from 'n8n-workflow';
+import {
+	validateNodeParameters,
+	type IExecuteFunctions,
+	type INodeExecutionData,
+} from 'n8n-workflow';
 
 import type { Content, GenerateContentResponse } from './interfaces';
 import { downloadFile, uploadFile } from './utils';
@@ -15,7 +19,11 @@ export async function baseAnalyze(
 	const text = this.getNodeParameter('text', i, '') as string;
 	const simplify = this.getNodeParameter('simplify', i, true) as boolean;
 	const options = this.getNodeParameter('options', i, {});
-
+	validateNodeParameters(
+		options,
+		{ maxOutputTokens: { type: 'number', required: false } },
+		this.getNode(),
+	);
 	const generationConfig = {
 		maxOutputTokens: options.maxOutputTokens,
 	};

packages/@n8n/nodes-langchain/nodes/vendors/GoogleGemini/helpers/interfaces.ts

@@ -1,4 +1,44 @@
+import type {
+	GenerateContentConfig,
+	GenerationConfig,
+	GenerateContentParameters,
+} from '@google/genai';
 import type { IDataObject } from 'n8n-workflow';
+export { Modality } from '@google/genai';
+
+/* type created based on: https://ai.google.dev/api/generate-content#generationconfig */
+export type GenerateContentGenerationConfig = Pick<
+	GenerationConfig,
+	| 'stopSequences'
+	| 'responseMimeType'
+	| 'responseSchema'
+	| 'responseJsonSchema'
+	| 'responseModalities'
+	| 'candidateCount'
+	| 'maxOutputTokens'
+	| 'temperature'
+	| 'topP'
+	| 'topK'
+	| 'seed'
+	| 'presencePenalty'
+	| 'frequencyPenalty'
+	| 'responseLogprobs'
+	| 'logprobs'
+	| 'speechConfig'
+	| 'thinkingConfig'
+	| 'mediaResolution'
+>;
+
+/* Type created based on: https://ai.google.dev/api/generate-content#method:-models.streamgeneratecontent */
+export interface GenerateContentRequest extends IDataObject {
+	contents: GenerateContentParameters['contents'];
+	tools?: GenerateContentConfig['tools'];
+	toolConfig?: GenerateContentConfig['toolConfig'];
+	systemInstruction?: GenerateContentConfig['systemInstruction'];
+	safetySettings?: GenerateContentConfig['safetySettings'];
+	generationConfig?: GenerateContentGenerationConfig;
+	cachedContent?: string;
+}
 
 export interface GenerateContentResponse {
 	candidates: Array<{

packages/@n8n/nodes-langchain/package.json

@@ -168,9 +168,9 @@
     "@azure/identity": "4.3.0",
     "@getzep/zep-cloud": "1.0.12",
     "@getzep/zep-js": "0.9.0",
-    "@google-ai/generativelanguage": "2.6.0",
     "@google-cloud/resource-manager": "5.3.0",
     "@google/generative-ai": "0.21.0",
+    "@google/genai": "1.19.0",
     "@huggingface/inference": "4.0.5",
     "@langchain/anthropic": "catalog:",
     "@langchain/aws": "0.1.11",