You got that right… it’s calling to gpt-4 turbo for personalized recommendations text. So it recommends movies, books, etc. One call to get recommendations than another call per rec to get personalized synopsis text. I tried making it 2 big calls but it was surprisingly slower. Lmk if thoughts
twice as fast is still pretty slow, what does it do? why is it so slow?
You got that right… it’s calling to gpt-4 turbo for personalized recommendations text. So it recommends movies, books, etc. One call to get recommendations than another call per rec to get personalized synopsis text. I tried making it 2 big calls but it was surprisingly slower. Lmk if thoughts
Making calls to GTP-4 will be pretty slow