Files
ik_llama.cpp/ggml
Iwan Kawrakow 09c94ae948 iq3_k: slightly faster Metal dequantize kernel
PP-512 goes to 473 t/s up from 452 t/s.
2024-07-31 08:49:44 +02:00
..
2024-07-27 07:55:01 +02:00
2024-07-30 16:11:25 +03:00
2024-07-27 07:55:01 +02:00
2024-07-27 07:55:01 +02:00