Files
ik_llama.cpp/ggml
Iwan Kawrakow e01045b02e iq4_kss: Metal
PP is not too bad - just 10% slower than q4_0.
But TG is 30% slower, i.e., predictably bad.
2024-10-16 14:14:00 +03:00
..
2024-07-27 07:55:01 +02:00
2024-10-16 14:14:00 +03:00
2024-10-16 14:14:00 +03:00
2024-07-27 07:55:01 +02:00
2024-10-04 14:43:26 +03:00