GPT-3.5 Turbo which debuted earlier this year is ideal for use cases that aren’t chat-specific. It can handle 4,000 tokens at a time which is double what previously offered models could interpret.
