feat(azure): update default provider to use responses API

.changeset/clever-toes-swim.md

@@ -0,0 +1,5 @@
+---
+'@ai-sdk/azure': patch
+---
+
+Update default Azure Provider to use Responses API

content/providers/01-ai-sdk-providers/04-azure.mdx

@@ -123,6 +123,11 @@ const enhancedModel = wrapLanguageModel({
 
 You can then use that enhanced model in functions like `generateText` and `streamText`.
 
+<Note>
+  The Azure provider calls the Responses API by default (unless you specify e.g.
+  `azure.chat`).
+</Note>
+
 ### Example
 
 You can use OpenAI language models to generate text with the `generateText` function:

examples/ai-core/src/generate-text/azure-responses-default.ts

@@ -0,0 +1,21 @@
+import 'dotenv/config';
+import { createAzure } from '@ai-sdk/azure';
+import { generateText } from 'ai';
+
+const azureDefault = createAzure({
+  fetch: async (input, init) => {
+    console.log('Azure  request URL:', input);
+    return fetch(input, init);
+  },
+});
+
+async function main() {
+  const result = await generateText({
+    model: azureDefault('gpt-5-nano'),
+    prompt: 'Write a short poem about the sea.',
+  });
+
+  console.log(result.text);
+}
+
+main().catch(console.error);

packages/azure/src/azure-openai-provider.test.ts

@@ -37,6 +37,118 @@ const server = createTestServer({
     {},
 });
 
+describe('responses (default language model)', () => {
+  describe('doGenerate', () => {
+    function prepareJsonResponse({
+      content = '',
+      usage = {
+        input_tokens: 4,
+        output_tokens: 30,
+        total_tokens: 34,
+      },
+    } = {}) {
+      server.urls[
+        'https://test-resource.openai.azure.com/openai/v1/responses'
+      ].response = {
+        type: 'json-value',
+        body: {
+          id: 'resp_67c97c0203188190a025beb4a75242bc',
+          object: 'response',
+          created_at: 1741257730,
+          status: 'completed',
+          model: 'test-deployment',
+          output: [
+            {
+              id: 'msg_67c97c02656c81908e080dfdf4a03cd1',
+              type: 'message',
+              status: 'completed',
+              role: 'assistant',
+              content: [
+                {
+                  type: 'output_text',
+                  text: content,
+                  annotations: [],
+                },
+              ],
+            },
+          ],
+          usage,
+          incomplete_details: null,
+        },
+      };
+    }
+
+    it('should set the correct default api version', async () => {
+      prepareJsonResponse();
+
+      await provider('test-deployment').doGenerate({
+        prompt: TEST_PROMPT,
+      });
+
+      expect(
+        server.calls[0].requestUrlSearchParams.get('api-version'),
+      ).toStrictEqual('v1');
+    });
+
+    it('should set the correct modified api version', async () => {
+      prepareJsonResponse();
+
+      await providerApiVersionChanged('test-deployment').doGenerate({
+        prompt: TEST_PROMPT,
+      });
+
+      expect(
+        server.calls[0].requestUrlSearchParams.get('api-version'),
+      ).toStrictEqual('2025-04-01-preview');
+    });
+
+    it('should pass headers', async () => {
+      prepareJsonResponse();
+
+      const provider = createAzure({
+        resourceName: 'test-resource',
+        apiKey: 'test-api-key',
+        headers: {
+          'Custom-Provider-Header': 'provider-header-value',
+        },
+      });
+
+      await provider('test-deployment').doGenerate({
+        prompt: TEST_PROMPT,
+        headers: {
+          'Custom-Request-Header': 'request-header-value',
+        },
+      });
+
+      expect(server.calls[0].requestHeaders).toStrictEqual({
+        'api-key': 'test-api-key',
+        'content-type': 'application/json',
+        'custom-provider-header': 'provider-header-value',
+        'custom-request-header': 'request-header-value',
+      });
+      expect(server.calls[0].requestUserAgent).toContain(
+        `ai-sdk/azure/0.0.0-test`,
+      );
+    });
+
+    it('should use the baseURL correctly', async () => {
+      prepareJsonResponse();
+
+      const provider = createAzure({
+        baseURL: 'https://test-resource.openai.azure.com/openai',
+        apiKey: 'test-api-key',
+      });
+
+      await provider('test-deployment').doGenerate({
+        prompt: TEST_PROMPT,
+      });
+      expect(server.calls[0].requestUrl).toStrictEqual(
+        'https://test-resource.openai.azure.com/openai/v1/responses?api-version=v1',
+      );
+    });
+  });
+});
+
 describe('chat', () => {
   describe('doGenerate', () => {
     function prepareJsonResponse({ content = '' }: { content?: string } = {}) {
@@ -72,7 +184,7 @@ describe('chat', () => {
     it('should set the correct default api version', async () => {
       prepareJsonResponse();
 
-      await provider('test-deployment').doGenerate({
+      await provider.chat('test-deployment').doGenerate({
         prompt: TEST_PROMPT,
       });
 
@@ -84,7 +196,7 @@ describe('chat', () => {
     it('should set the correct modified api version', async () => {
       prepareJsonResponse();
 
-      await providerApiVersionChanged('test-deployment').doGenerate({
+      await providerApiVersionChanged.chat('test-deployment').doGenerate({
         prompt: TEST_PROMPT,
       });
 
@@ -104,7 +216,7 @@ describe('chat', () => {
         },
       });
 
-      await provider('test-deployment').doGenerate({
+      await provider.chat('test-deployment').doGenerate({
         prompt: TEST_PROMPT,
         headers: {
           'Custom-Request-Header': 'request-header-value',
@@ -130,7 +242,7 @@ describe('chat', () => {
         apiKey: 'test-api-key',
       });
 
-      await provider('test-deployment').doGenerate({
+      await provider.chat('test-deployment').doGenerate({
         prompt: TEST_PROMPT,
       });
       expect(server.calls[0].requestUrl).toStrictEqual(

packages/azure/src/azure-openai-provider.ts

@@ -28,7 +28,7 @@ export interface AzureOpenAIProvider extends ProviderV3 {
   (deploymentId: string): LanguageModelV3;
 
   /**
-Creates an Azure OpenAI chat model for text generation.
+  Creates an Azure OpenAI responses API model for text generation.
    */
   languageModel(deploymentId: string): LanguageModelV3;
 
@@ -239,11 +239,11 @@ export function createAzure(
       );
     }
 
-    return createChatModel(deploymentId);
+    return createResponsesModel(deploymentId);
   };
 
   provider.specificationVersion = 'v3' as const;
-  provider.languageModel = createChatModel;
+  provider.languageModel = createResponsesModel;
   provider.chat = createChatModel;
   provider.completion = createCompletionModel;
   provider.embedding = createEmbeddingModel;