llama.cpp
CUDA: faster dequantize kernels for Q4_0 and Q4_1
#4938

Merged

CUDA: faster dequantize kernels for Q4_0 and Q4_1 #4938

ikawrakow merged 1 commit into master from ik/cuda_faster_legacy_dequantize

CUDA: faster dequantize kernels for Q4_0 and Q4_1

08b89f7e

JohannesGaessler commented on 2024-01-14

JohannesGaessler approved these changes on 2024-01-14

ikawrakow merged 4a3156de into master 2 years ago

ikawrakow deleted the ik/cuda_faster_legacy_dequantize branch 2 years ago

ggerganov commented on 2024-01-15

Reviewers

JohannesGaessler

ggerganov

Assignees

No one assigned

Labels

None yet

Milestone

No milestone