Hacker News new | ask | show | jobs
by tjake 1042 days ago
The default for `convert.py` is F32. This is just SIMD CPU comparison.

Jlama uses the vector api in java20 but also better thread scheduling with work stealing and zero allocation.
2 comments
belfthrow 1041 days ago
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.
link
version_five 1042 days ago
Very interesting, I'll watch for the quantized version.
link