I think all of us have heard about DeepSeek’s R1 model which was released recently. It’s claimed that it offers O1 performance for a fraction of the cost. Can’t wait to try it in Cursor!
Or can we use our own api keys for a model like that with cursor? Not sure if this is available.
every body wants this but i don’t know why they are not able to catch the public pulse,amd they are not officially telling that they’re cooking it!
Holy smokes I also have it!
Why is R1 so much slower than the official one?
But the effect is very, very, very accurate
Hey, we’re using the API through our provider, Fireworks, not directly from DeepSeek. They run the model on their own equipment, which ensures Cursor’s privacy and security remains unchanged, but does explain why the speed is significantly lower. I believe they’ll add more capacity soon, and the model will noticeably speed up.
Are you trying to use this model in Chat or Composer?
It looks like you have a dot in the model name, it has to be an exact match with the API you are using.
Want for both composer and chat
DeepSeek R1 will work in the chat via an API key already, but we are looking at allowing Composer support too, as long as you have a Pro subscription.
Oh, great to know, was very doubtful of using Deepseek for privacy reasons, lol