Really excited about the new precision finding remote.

Will this strategy work on lower memory unified memory machines like Apple Silicon or AMD?

There are solutions already for Mac , check out ISTA DAs lab hugginface, I got one of their qwen next models working pretty fast on m2 ultra

I do feel since I use hermes on my work windows computer and grok bot in the cloud on everything else that computer use is so useful and important, but I'm also worried that Apple will not be a good choice for that.

The problem I have is that when I ssh into my mac, I want to be able to control software from the command line and having a popup authorization window that just halts the software being controlled is not good for me. I'm okay with having the authorization, but they need to make it visible to ssh users when ssh'ing in.

did this come from the ideocracy universe?

no one wants to use them anyways. i pay $100 a month for ultra and i'm using claude and chatgpt.

i would really like to see something like this for images.

doesn't using vision instead of notation burn tokens?

I would really like something similar for local ai where the x axis is vram.

I'm really looking for a multi-modal image capable version of Jev.

If we could get machine learning type results on images without training, that would be fantastic.

Jev’s context should have signals/features to operate on. Same way LLMs can use CV & code to analyze an image.