The neural engine is perfectly capable of accelerating matmults. It's just that autoregressive decoding in single batch LLM inference is memory bandwidth constrained, so there are no performance benefits to using the ANE for LLM inference (although, there's a huge power efficiency benefit). And the only way to use the neural engine is via CoreML. Using the GPU with MLX or MPS is often easier.

I have to assume they’re doing something like that in the lab for 4 years from now.