ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-02-20 21:24:08 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
b7768e203f3283d601c9d3a0cd34dfc76b40aa87
ik_llama.cpp/ggml/src
History
Kawrakow b7768e203f Faster CPU prompt processing for Q4_K and Q5_K (#525) 
* q4_K: dequantize to q8_1_r8 for batch >= 32

We get 268 t/s, up from 186 t/s.

* q4_K: GEMM with q8_2_X4

* q5_K: GEMM with q8_2_X4 and repack to q8_1_r8

* Remove the scales, they are not needed

---------

Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
2025-06-13 07:58:15 +03:00
..
ggml-cann
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-cuda
IQ1_M_R4 CUDA implementation (#494)
2025-06-05 19:13:51 +03:00
ggml-sycl
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
iqk
Faster CPU prompt processing for Q4_K and Q5_K (#525)
2025-06-13 07:58:15 +03:00
kompute @ 4565194ed7
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
kompute-shaders
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
llamafile
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
vulkan-shaders
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
CMakeLists.txt
Better strategy for GPU offload (#520)
2025-06-12 19:25:11 +03:00
ggml-aarch64.c
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-aarch64.h
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
ggml-alloc.c
Fix ARM_NEON build failure due to q8_2 (#303)
2025-04-01 13:48:20 +02:00
ggml-backend-impl.h
Merge mainline llama.cpp (#3)
2024-07-27 07:55:01 +02:00
ggml-backend.c
Fix non rpc build error (#506)
2025-06-08 17:27:00 +03:00
ggml-blas.cpp
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-cann.cpp
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-common.h
Trellis quants with CPU inference (#441)
2025-05-23 09:17:52 +03:00
ggml-cuda.cu
Better strategy for GPU offload (#520)
2025-06-12 19:25:11 +03:00
ggml-impl.h
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-kompute.cpp
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-metal.m
Metal implementatio for the trellis quants. (#475)
2025-06-01 15:23:44 +03:00
ggml-metal.metal
Metal implementatio for the trellis quants. (#475)
2025-06-01 15:23:44 +03:00
ggml-quants.c
Trellis quants with CPU inference (#441)
2025-05-23 09:17:52 +03:00
ggml-quants.h
IQ1_M_R4: better 1.75 bpw quants (#187)
2025-02-06 14:08:52 +02:00
ggml-rpc.cpp
Fix non rpc build error (#506)
2025-06-08 17:27:00 +03:00
ggml-sycl.cpp
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml-vulkan.cpp
Merge mainline - Aug 12 2024 (#17)
2024-08-12 15:14:32 +02:00
ggml.c
Faster CPU prompt processing for Q4_K and Q5_K (#525)
2025-06-13 07:58:15 +03:00