Skip to content

feat: add openai responses api capability #7

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

feat: add openai responses api capability #7

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
00a1f37
chore: update dependencies
matthewmichel Mar 12, 2025
23235ec
chore: use openai responses api
matthewmichel Mar 12, 2025
c7b6d04
feat: add web search tool example
matthewmichel Mar 12, 2025
ce39748
chore: align naming conventions
matthewmichel Mar 12, 2025
d74b983
chore: disable storing responses to openai
matthewmichel Mar 12, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 11 additions & 1 deletion app/(chat)/api/chat/route.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@ import { getWeather } from '@/lib/ai/tools/get-weather';
import { isProductionEnvironment } from '@/lib/constants';
import { NextResponse } from 'next/server';
import { myProvider } from '@/lib/ai/providers';
import { openai } from '@ai-sdk/openai';

export const maxDuration = 60;

Expand Down Expand Up @@ -85,6 +86,7 @@ export async function POST(request: Request) {
'createDocument',
'updateDocument',
'requestSuggestions',
'webSearchPreview'
],
experimental_transform: smoothStream({ chunking: 'word' }),
experimental_generateMessageId: generateUUID,
Expand All @@ -96,6 +98,14 @@ export async function POST(request: Request) {
session,
dataStream,
}),
webSearchPreview: openai.tools.webSearchPreview({
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Configuration Management Issue: Hardcoded configuration values ('medium' for search context size) directly in the route handler reduce modifiability and increase technical debt.

Consider extracting configuration values to constants or configuration files:

// In a configuration file or constants file
export const AI_CONFIG = {
  webSearch: {
    contextSize: 'medium',
  },
  openai: {
    store: false,
  }
};

// Then reference in the handler
webSearchPreview: openai.tools.webSearchPreview({
  searchContextSize: AI_CONFIG.webSearch.contextSize
}),

searchContextSize: 'medium'
})
},
providerOptions: {
openai: {
store: false,
}
},
onFinish: async ({ response, reasoning }) => {
if (session.user?.id) {
Expand Down Expand Up @@ -134,7 +144,7 @@ export async function POST(request: Request) {
});
},
onError: () => {
return 'Oops, an error occured!';
return 'Oops, an error occurred!';
},
});
} catch (error) {
Expand Down
4 changes: 2 additions & 2 deletions lib/ai/providers.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,8 +25,8 @@ export const myProvider = isTestEnvironment
})
: customProvider({
languageModels: {
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Architecture Pattern Issue: Inconsistent provider initialization pattern between the new OpenAI responses API and existing providers. This creates an inconsistent mental model for developers and could lead to confusion during maintenance.

Consider using a consistent initialization pattern across all providers:

// Either update all providers to use the new pattern
'chat-model-small': openai.responses('gpt-4o-mini'),
'chat-model-large': openai.responses('gpt-4o'),

// Or maintain the previous pattern if it's still supported
'chat-model-small': openai('gpt-4o-mini'),
'chat-model-large': openai('gpt-4o'),

'chat-model-small': openai('gpt-4o-mini'),
'chat-model-large': openai('gpt-4o'),
'chat-model-small': openai.responses('gpt-4o-mini'),
'chat-model-large': openai.responses('gpt-4o'),
'chat-model-reasoning': wrapLanguageModel({
model: fireworks('accounts/fireworks/models/deepseek-r1'),
middleware: extractReasoningMiddleware({ tagName: 'think' }),
Expand Down
4 changes: 2 additions & 2 deletions package.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@
},
"dependencies": {
"@ai-sdk/fireworks": "0.1.12",
"@ai-sdk/openai": "1.2.0",
"@ai-sdk/openai": "1.2.2",
"@ai-sdk/react": "^1.1.20",
"@codemirror/lang-javascript": "^6.2.2",
"@codemirror/lang-python": "^6.1.6",
Expand All @@ -40,7 +40,7 @@
"@vercel/analytics": "^1.3.1",
"@vercel/blob": "^0.24.1",
"@vercel/postgres": "^0.10.0",
"ai": "4.1.50",
"ai": "4.1.54",
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Verification agent

🧩 Analysis chain

Package update to [email protected]

This update from 4.1.50 to 4.1.54 is required to support the OpenAI Responses API implementation. The PR comments indicate this adds 14 new transitive dependencies, which is expected for this feature addition.

🏁 Script executed:

#!/bin/bash
# Verify the compatibility of these packages by checking their latest versions
echo "Checking latest versions of @ai-sdk/openai and ai packages..."
npm view @ai-sdk/openai version
npm view ai version

Length of output: 202

Update ai dependency to the latest published version (4.3.10)
npm registry shows that ai@4.1.54 is two minor releases behind—the current latest is 4.3.10. To pick up bug fixes and improvements beyond the OpenAI Responses API support, please bump the version.

  • File: package.json (line 43)
  • Diff suggestion:
-   "ai": "4.1.54",
+   "ai": "^4.3.10",
📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change
"ai": "4.1.54",
"ai": "^4.3.10",

"bcrypt-ts": "^5.0.2",
"class-variance-authority": "^0.7.0",
"classnames": "^2.5.1",
Expand Down
97 changes: 81 additions & 16 deletions pnpm-lock.yaml
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.