]> git.djapps.eu Git - pkg/ggml/sources/ggml/commit
CUDA: faster dequantize kernels for Q4_0 and Q4_1 (llama/4938)
authorKawrakow <redacted>
Mon, 15 Jan 2024 05:48:06 +0000 (07:48 +0200)
committerGeorgi Gerganov <redacted>
Wed, 17 Jan 2024 18:44:10 +0000 (20:44 +0200)
commita00a9f28a49a16223eddf5178ac3e694cd2df611
tree8b7207dcbf9a236875528717cc55b523980cfa27
parent327eb4b67c116ae11c1ead0d7c57f0abdac227ce
CUDA: faster dequantize kernels for Q4_0 and Q4_1 (llama/4938)

Co-authored-by: Iwan Kawrakow <redacted>
src/ggml-cuda.cu