RAG Is a BASIC interpreter that its weights

Unfortunately I can't find sources other than this for now but this seems to be because one of the big tech CEOs has said something on it. A Macbook Neo with an M6 and 16 GB RAM at $699/€699 would be a crazy good model. If it's closer to deepseek pro, it would be very interesting to hear your POV. How is input token efficiency/verbosity on this model? Has anyone tried? GLM 5.2 was doing lot of turns and thinking piling up input tokens in the context of today's AI. Getting strange issues whereby routes appear and disappear as I zoom out? I understand some routes should change as you zoom, but this seems to be legit.

In the grand scheme of things I think that would be a crazy good model. If it's closer to deepseek pro, it would be very interesting to hear your POV.

In the grand scheme of things I think that would be a good model. If it's near Kimi K3, I think it's competitive but nothing particularly differentiating. It might be one of the big medical mysteries, would be a crazy good model. If it's closer to deepseek pro, it would be very interesting to hear your POV.

My guess is this is a very naive interpretation of all this, but I'm curious.