]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: faster q8_0 -> f16 dequantization (llama/4895)
authorJohannes Gäßler <redacted>
Fri, 12 Jan 2024 19:38:54 +0000 (20:38 +0100)
committerGeorgi Gerganov <redacted>
Sat, 13 Jan 2024 22:06:46 +0000 (00:06 +0200)
commitbcdb75e39621690aa50da42343aee27e6624c994
tree9cb875f0f15ee761c37363b1a833534b4817d952
parent400c07f00508e6f60fb25405444b5669c365b0a9
CUDA: faster q8_0 -> f16 dequantization (llama/4895)
src/ggml-cuda.cu