Is Cursor going to get an update to the latest GPT-4
GPT-4 Turbo is more capable and has knowledge of world events up to April 2023
GPT-4 Turbo at a 3x cheaper price for input tokens and a 2x cheaper price for output tokens compared to GPT-4
Blog Post by OpenAI - New models and developer products announced at DevDay
Hi!, I am a fellow browser of the forum and I am aware you are excited! and relate to it but that is what forums are for, you could have searched about this, they have already planned this and will be releasing one shortly either today or tmr
What I’m most excited for is that 128k context window. Cursor will be able to see massively more information before generating a response
Where is this announcement? I couldn’t find it
Thank you for the update and for working urgently on it!
Regarding the code smarts, according to OpenAI, the new 3.5 (not 4) Turbo supports improved instruction following, JSON mode, reproducible outputs, parallel function calling. “For instance, our internal evals show a 38% improvement on format following tasks such as generating JSON, XML and YAML”.
So I’m optimistic on this
About the rate limits and capacity, what about a provisional implementation of GPT-4 Turbo capped at 8K, so that we/you could start to enjoy the speed and cost advantage of it?
Finally, a clarification: What happens today if I set Cursor to use my OpenAI API key? I can’t see any details about which model it uses and how many tokens/$ I spend each use, as I’m used to seeing and appreciating, for example, in Typing Mind.
Furthermore, I just tried to use my API key with v0.15.1 and I surprisingly get this message, while it works with GPT-4 Turbo on Typing Mind: