I don’t really get how rate limits work. I’m just about to register a company as I’m building a webapp around a great idea I have for AI that I think will really succeed. My problem though is rate limit. Firstly looking at the rate limit at openAI developer docs it doesn’t even mention gpt-3.5-turbo which is the model I want to use. But the link to gptforwork.com does. But it states that after 48 hours the rate limit is 3500 requests per minute for gpt-3.5-turbo. But it says “davinci tokens”, and davinci is 10x the price over gpt-3.5-turbo, which is what I want to use. This is probably a stupid quest, but yeah any help here would be awesome. Does it mean that after 48 hours of using payed gpt-3.5-turbo the max requests per minute is 3500? Also, the davinci tokens? Does it mean that the 3500/90k tokens is just if I use davinci with gpt-3.5-turbo? Or is it 3500 requests per minute for gpt-3.5-turbo and another number than 90K tokens for gpt-3.5-turbo? If so, what is that token number?
Also can a gpt-3,5-turbo model be fine tuned and trained? Or is that other models?
Sorry for probably sounding super stupid, but yeah I’m new to this and very confused. I’ve googled this for hours but I don’t find any conclusive answers.