I was hoping for an on-device LLM from Apple, but this might be fine:

I was hoping for an on-device LLM from Apple, but this might be fine:

https://appleinsider.com/articles/24/05/30/apple-and-openai-allegedly-reach-deal-to-bring-chatgpt-functionality-to-ios-18

👨‍💻 /uno and /outside, @unofficial

We’re gonna need them to go ahead and get started then

probably would take a year to do a first pass at embedding on device haha

On-device LLM trained specifically to you would amazing, hopefully in time

I would really like embedded models, but I think Apple is way too far behind to ship that this cycle. Maybe next year, we'll get embedded models with custom ASICs for inference.