llama.cpp
Implementations for Q4_0_8_8 quantization based functions in AVX2 SIMD architecture
#8713
Merged
Overview
Commits
4
Changes
View On GitHub

Implementations for Q4_0_8_8 quantization based functions in AVX2 SIMD architecture #8713

ggerganov merged 4 commits into ggml-org:master from Srihari-mcw:block_interleaving_q4_0_8_8_avx2_implementation
Srihari-mcw
github-actions github-actions added ggml
Srihari-mcw
Srihari-mcw commented on 2024-07-26
bartowski1182
Srihari-mcw
nisten
mofosyne mofosyne added Review Complexity : Medium
ggerganov
ggerganov
ggerganov approved these changes on 2024-09-03
Srihari-mcw Add AVX2 based implementations for quantize_q8_0_4x8, ggml_gemv_q4_0_…
0c81b7bc
Srihari-mcw Update code to fix issues occuring due to non alignment of elements t…
49af3f5d
Srihari-mcw Update comments and indentation
364dc964
Srihari-mcw Make updates to reduce number of load instructions
c950fc30
Srihari-mcw Srihari-mcw force pushed to c950fc30 2 years ago
Srihari-mcw
ggerganov ggerganov merged 581c3051 into master 2 years ago
slaren

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Review Complexity : Medium ggml
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum