4 ms·
No, my experience is that GPT-3.5-turbo generates markedly poorer results than text-davinci-003 and -002.
by npsomaratna 3y ago
No, my experience is that GPT-3.5-turbo generates markedly poorer results than text-davinci-003 and -002.
- qeternity 3y agoAgreed, it's general consensus that 3.5-turbo is a substantially weaker model.