2 ms·
> that support fast and lossless inference of 1.58-bit models on CPU (with NPU and GPU support coming next).
by diggan 2y ago
> that support fast and lossless inference of 1.58-bit models on CPU (with NPU and GPU support coming next).
loading thread…