]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
metal : add Q8_0 support (#2763)
authorGeorgi Gerganov <redacted>
Thu, 24 Aug 2023 13:19:57 +0000 (16:19 +0300)
committerGitHub <redacted>
Thu, 24 Aug 2023 13:19:57 +0000 (16:19 +0300)
commitd67777c202c03bcb74372690599ef3c03affb3ba
tree13ac15fdd3688c4b1468018cc408260b9ca2c4aa
parentc3e53b421a9910548be0345f85712c535f467a98
metal : add Q8_0 support (#2763)

* metal : add dequantize_q8_0 kernel

* metal : add mul_mat_q8_0_f32 kernel

* metal : add Q8_0 mul_mm kernel
ggml-metal.m
ggml-metal.metal