llama.cpp
ggml : alternative Q4_3 implementation using modified Q8_0
#1109

Merged

ggml : alternative Q4_3 implementation using modified Q8_0 #1109

ggerganov merged 5 commits into master from q4_3b

ggerganov marked this pull request as ready for review 2 years ago

sw commented on 2023-04-21

ggml : prefer vzip to vuzp

ec805eef

ggml : alternative Q4_3 implementation using modified Q8_0

5425e060

ggml : fix Q4_3 scalar imlpementation

829c4806

ggml : slight improvement of Q4_3 - no need for loop unrolling

76b6b267

ggerganov force pushed to 76b6b267 2 years ago

ggml : fix AVX paths for Q8_0 quantization

2c358eca

ggerganov merged 955ef9a5 into master 2 years ago

ggerganov deleted the q4_3b branch 2 years ago

Reviewers

Assignees

No one assigned

Labels

None yet

Milestone

No milestone