nikola/llama_cpp
1
0
Fork 0
Code Issues Pull Requests Actions 33 Packages Projects Releases Wiki Activity
3,688 Commits 1 Branch 0 Tags
fbb7fcffbcfc50009c275e75982b1603a6cb6b80
Commit Graph
4 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Srihari-mcwandGitHub 581c305186 ggml : AVX2 support for Q4_0_8_8 (#8713) 
* Add AVX2 based implementations for quantize_q8_0_4x8, ggml_gemv_q4_0_8x8_q8_0 and ggml_gemm_q4_0_8x8_q8_0 functions

* Update code to fix issues occuring due to non alignment of elements to be processed as multiple of 16 in MSVC

* Update comments and indentation

* Make updates to reduce number of load instructions
 2024-09-04 19:51:22 +03:00
Srihari-mcwandGitHub ea5d7478b1 sgemm : improved Q4_0 and Q8_0 performance via 4xN and Mx4 gemm (#8908) 2024-08-31 11:20:35 +03:00
Srihari-mcwandGitHub 5b6da18750 Make updates to type cast based on compiler instead of OS (#7851) 2024-06-17 20:23:17 +02:00
Srihari-mcwandGitHub 33c8d50acc Add provisions for windows support for BF16 code including CMake provision for enabling AVX512_BF16 (#7258) 2024-05-20 12:18:39 +10:00