Also, it’s possible to do inference on less than cutting edge hardware. The open models will propagate allover the world and people will run their own local, fi

Also, it’s possible to do inference on less than cutting edge hardware. The open models will propagate allover the world and people will run their own local, fine tuned models on e waste.