Where does the performance difference come from? And in what kind of processor & gpu? I didn't even know llama.cpp had a 32 bit option. For now I'm pretty suspicious it's a fair comparison.
Could you link to some of the examples in your repo where you enforce the zero allocation? I don't see much reuse of the buffers, eg float buffers and there is quite a lot of array based heap allocation. Just for my own interest. Many thanks. Cool to see the use of the new vector api also.
Jlama uses the vector api in java20 but also better thread scheduling with work stealing and zero allocation.