Node.js OpenAI library v4.20.1 constantly timing out on basic requests

[jackw]

I’m experiencing ongoing timeout issues when trying to use OpenAI’s API with the Node.js library version 4.20.1. Even for very simple queries, like basic math questions, I keep running into timeouts.

Here’s the implementation I’m currently working with:

const { OpenAI } = require('openai');

async function handleChatRequest(request, response) {
  try {
    const client = new OpenAI({
      apiKey: process.env.OPENAI_SECRET_KEY,
    });

    const userInput = request.body.message;

    const completion = await client.chat.completions.create({
      model: "gpt-4", 
      messages: [{ role: 'user', content: userInput }],
    });

    response.json(completion.choices);
  } catch (err) {
    console.log('OpenAI request failed:', err);
    response.status(500).json({ error: 'Request failed' });
  }
}

exports.handleChatRequest = handleChatRequest;

What am I missing? I’m attempting to create a simple ChatGPT clone using the official OpenAI API, but these timeout errors are really hindering progress. Any suggestions on what might be going wrong?

[Tom01_Wonder]

I encountered similar timeout issues with the Node.js library version 4.20.1. A common solution is to explicitly set a timeout in the client configuration. GPT-4 tends to take significantly more time for responses than GPT-3.5, so increasing the timeout may resolve the problem. For instance, you can set it to 60 seconds as follows:

const client = new OpenAI({
  apiKey: process.env.OPENAI_SECRET_KEY,
  timeout: 60000, // 60 seconds
});

Additionally, ensure your internet connection is stable and monitor the rate limits. I temporarily switched to gpt-3.5-turbo while troubleshooting, which helped pinpoint the cause of the timeouts. It’s also beneficial to log the complete error response to gain more insight into the timeout issues.