Switch to OpenAI Responses API (#1981)

## Summary

#1960 added support for OpenAI's chat completions api.

This change switches to using [OpenAI's new Responses API](https://developers.openai.com/api/docs/guides/migrate-to-responses) instead.

### How to test locally or on Vercel

### How to test locally
1. Set env vars:
`AI_PROVIDER=openai AI_API_KEY= AI_BASE_URL=<> AI_MODEL_NAME=<> AI_REQUEST_HEADERS={"X-Client-Id":"","X-Username":"", AI_ADDITIONAL_OPTIONS = {API_TYPE: "responses"}}`
3. Open Hyperdx's chart explorer and use the AI assistant chart builder
   - e.g. "show me error count by service in the last hour"
4. Confirm the assistant returns a valid chart config.

### References



- Linear Issue:
- Related PRs:
Co-authored-by: peter-leonov-ch <209667683+peter-leonov-ch@users.noreply.github.com>

Author: vinzee

.changeset/sharp-eggs-appear.md

@@ -0,0 +1,5 @@
+---
+'@hyperdx/api': patch
+---
+
+Update OpenAI model configuration to use the new Responses API

packages/api/src/controllers/__tests__/ai.test.ts

@@ -13,9 +13,11 @@ const mockCreateAnthropic = jest.fn(
   (_opts?: Record<string, unknown>) => mockAnthropicFactory,
 );
 
-const mockOpenAIChatFactory = jest.fn((_model?: string) => mockOpenAIModel);
+const mockOpenAIResponsesFactory = jest.fn(
+  (_model?: string) => mockOpenAIModel,
+);
 const mockCreateOpenAI = jest.fn((_opts?: Record<string, unknown>) => ({
-  chat: mockOpenAIChatFactory,
+  responses: mockOpenAIResponsesFactory,
 }));
 
 jest.mock('@ai-sdk/anthropic', () => ({
@@ -191,7 +193,7 @@ describe('openai provider', () => {
     expect(mockCreateOpenAI).toHaveBeenCalledWith(
       expect.objectContaining({ apiKey: 'sk-test' }),
     );
-    expect(mockOpenAIChatFactory).toHaveBeenCalledWith('gpt-4o');
+    expect(mockOpenAIResponsesFactory).toHaveBeenCalledWith('gpt-4o');
   });
 
   it('passes baseURL when AI_BASE_URL is set', () => {

packages/api/src/controllers/ai.ts

@@ -368,9 +368,7 @@ function getAnthropicModel(): LanguageModel {
 }
 
 /**
- * Configure OpenAI-compatible model.
- * Works with any OpenAI Chat Completions-compatible endpoint
- * (e.g. Azure OpenAI, OpenRouter, LiteLLM proxies).
+ * Configure OpenAI-compatible model using the Responses API (/v1/responses).
  */
 function getOpenAIModel(): LanguageModel {
   const apiKey = config.AI_API_KEY;
@@ -399,5 +397,5 @@ function getOpenAIModel(): LanguageModel {
     ...(Object.keys(headers).length > 0 && { headers }),
   });
 
-  return openai.chat(config.AI_MODEL_NAME);
+  return openai.responses(config.AI_MODEL_NAME);
 }