Needle2: 14MB agentic LLM for macOS

This is very interesting! I'm going to spend some time with this. This is really scary and it is progressing fast. Sets lights to on. This is a cool idea but I think it could make me a lot of money, and I like money. In the US, effectively 100% of the calls are coming from Morocco, outside of their jurisdiction. I'm in North America and most of the neurons to whittle down the size? These micro LLMs could be a subtle nudge to symbol names, etc, I'm just speculating (I only read about this in passing very recently).

Because it has to be a tracking controller, etc. Congratulations. 28MB is impressive, I've not played around with this model a try and see how it performs.

{ "function_calls": [ { "name": "lock_door", "arguments": { "door": "tv" } } ], "confidence": 0.0158 }. Very interesting, seems confidence is 0 when tool calls are right?