The model `davinci-3-5` does not exist

[livid ermine]

try {
const response = await axios.post('https://api.openai.com/v1/engines/davinci-3-5/completions', {
prompt: prompt,
max_tokens: 100,
n: 1,
temperature: 0.7,
stop: ['\n']
},

someone that can help me out?

[placid vector]

Because it doesn't. There's GPT-3 (text-davinci-003) & chatgpt(gpt-3.5-turbo). These are the 2 models.

[livid ermine]

@placid vector So it should look like this? try {
const response = await axios.post('https://api.openai.com/v1/engines/gpt-3.5-turbo/completions', {
prompt: prompt,
max_tokens: 100,
n: 1,
temperature: 0.7,
stop: ['\n']
}, {
headers: {
'Authorization': Bearer ${apiKey},
'Content-Type': 'application/json'
}
});

[livid ermine]

I get this problem know. AI model response: {
id: 'cmpl-73Stg7BIJ831nNrbHrHfcazDYta0R',
object: 'text_completion',
created: 1681060420,
model: 'davinci',
choices: [
{
text: '__________________',
index: 0,
logprobs: null,
finish_reason: 'stop'
}
],

Ahh. I need to train the model. LOL

[placid vector]

No.

You're calling the wrong endpoint(api)

Here you go for GPT DaVinci: https://platform.openai.com/docs/api-reference/completions/create

you need to call the model in the body(above prompt. And for chatgpt it's messages not prompt)

And this is for chatGPT: https://platform.openai.com/docs/api-reference/chat

[livid ermine]

So nothing like this?

const response = await axios.post('https://api.openai.com/v1/engines/text-davinci-002/completions', {

[placid vector]

Engines is deprecated. You need to use model

[livid ermine]

Thanks,

First time ever building a frontend and backend. So many questions 🙂

[placid vector]

that's a good thing, no problem 🙂