diff --git a/.changeset/fine-kiwis-rush.md b/.changeset/fine-kiwis-rush.md
new file mode 100644
index 000000000..b2460fabe
--- /dev/null
+++ b/.changeset/fine-kiwis-rush.md
@@ -0,0 +1,5 @@
+---
+'@modelcontextprotocol/client': patch
+---
+
+fix: validate sampling/createMessage with correct schema
diff --git a/packages/client/src/client/client.ts b/packages/client/src/client/client.ts
index 8d96ba0bc..82363cef2 100644
--- a/packages/client/src/client/client.ts
+++ b/packages/client/src/client/client.ts
@@ -43,6 +43,7 @@ import {
     CompleteResultSchema,
     CreateMessageRequestSchema,
     CreateMessageResultSchema,
+    CreateMessageResultWithToolsSchema,
     CreateTaskResultSchema,
     ElicitRequestSchema,
     ElicitResultSchema,
@@ -458,8 +459,9 @@ export class Client<
                     return taskValidationResult.data;
                 }
 
-                // For non-task requests, validate against CreateMessageResultSchema
-                const validationResult = safeParse(CreateMessageResultSchema, result);
+                // For non-task requests, validate against appropriate schema
+                const schema = params.tools ? CreateMessageResultWithToolsSchema : CreateMessageResultSchema;
+                const validationResult = safeParse(schema, result);
                 if (!validationResult.success) {
                     const errorMessage =
                         validationResult.error instanceof Error ? validationResult.error.message : String(validationResult.error);
diff --git a/test/integration/test/sampling_tools.test.ts b/test/integration/test/sampling_tools.test.ts
new file mode 100644
index 000000000..a98e47196
--- /dev/null
+++ b/test/integration/test/sampling_tools.test.ts
@@ -0,0 +1,104 @@
+import { Client } from '@modelcontextprotocol/client';
+import { CreateMessageRequestSchema, InMemoryTransport } from '@modelcontextprotocol/core';
+import { Server } from '@modelcontextprotocol/server';
+import { describe, expect, test } from 'vitest';
+
+describe('sampling/createMessage with tools', () => {
+    test('should support returning tool calls when tools are provided', async () => {
+        const server = new Server({ name: 'test server', version: '1.0' }, { capabilities: {} });
+
+        const client = new Client(
+            { name: 'test client', version: '1.0' },
+            {
+                capabilities: {
+                    sampling: {
+                        tools: {}
+                    }
+                }
+            }
+        );
+
+        // Implement request handler for sampling/createMessage that returns a tool call
+        client.setRequestHandler(CreateMessageRequestSchema, async _request => {
+            return {
+                model: 'test-model',
+                role: 'assistant',
+                stopReason: 'toolUse',
+                content: [
+                    {
+                        type: 'tool_use',
+                        id: 'call_1',
+                        name: 'test_tool',
+                        input: { arg: 'value' }
+                    }
+                ]
+            };
+        });
+
+        const [clientTransport, serverTransport] = InMemoryTransport.createLinkedPair();
+        await Promise.all([client.connect(clientTransport), server.connect(serverTransport)]);
+
+        const result = await server.createMessage({
+            messages: [{ role: 'user', content: { type: 'text', text: 'use the tool' } }],
+            maxTokens: 100,
+            tools: [{ name: 'test_tool', inputSchema: { type: 'object' } }]
+        });
+
+        expect(result).toEqual({
+            model: 'test-model',
+            role: 'assistant',
+            stopReason: 'toolUse',
+            content: [
+                {
+                    type: 'tool_use',
+                    id: 'call_1',
+                    name: 'test_tool',
+                    input: { arg: 'value' }
+                }
+            ]
+        });
+    });
+
+    test('should fail if returning tool calls when tools are NOT provided', async () => {
+        const server = new Server({ name: 'test server', version: '1.0' }, { capabilities: {} });
+
+        const client = new Client(
+            { name: 'test client', version: '1.0' },
+            {
+                capabilities: {
+                    sampling: {}
+                }
+            }
+        );
+
+        // Implement request handler for sampling/createMessage that returns a tool call
+        client.setRequestHandler(CreateMessageRequestSchema, async _request => {
+            return {
+                model: 'test-model',
+                role: 'assistant',
+                stopReason: 'toolUse',
+                content: [
+                    {
+                        type: 'tool_use',
+                        id: 'call_1',
+                        name: 'test_tool',
+                        input: { arg: 'value' }
+                    }
+                ]
+            };
+        });
+
+        const [clientTransport, serverTransport] = InMemoryTransport.createLinkedPair();
+        await Promise.all([client.connect(clientTransport), server.connect(serverTransport)]);
+
+        // This should fail because the client validation will reject the tool call result
+        // when validating against CreateMessageResultSchema (since tools were not requested)
+        await expect(
+            server.createMessage({
+                messages: [{ role: 'user', content: { type: 'text', text: 'use the tool' } }],
+                maxTokens: 100
+                // No tools provided
+            })
+        ).rejects.toThrow();
+    });
+});