Hey,
I thought this was supposed to be easy lol! Apparently I’m not sending the correct request? Doing this in React for a Next.js application if that matters.
Here’s a snippet of my code:
export default async function (req, res) {
if (!checkApiKey(res)) return;
const userMessage = req.body.userMessage || "";
// Log the received request body
console.log("Received request body:", req.body);
// Prepare the messages for the chat conversation
const inputs = [
{ role: "system", content: "You are a wise and loving grandmother." },
{ role: "user", content: userMessage },
];
try {
const response = await openai.createCompletion({
model: "gpt-3.5-turbo",
messages: inputs, // Use the messages parameter
max_tokens: 100,
temperature: 0.7,
});
A little confused why it’s not recognizing the messages input? This is exactly what the docs say to do here: OpenAI API
Error response data: {
error: {
message: 'Unrecognized request argument supplied: messages',
type: 'invalid_request_error',
param: null,
code: null
}
}