| .. |
|
dpct
|
8293970542
SYCL: Rename oneMKL to oneMath (#12192)
|
10 bulan lalu |
|
CMakeLists.txt
|
94148ba330
sycl: allow ggml-sycl configuration and compilation using Visual Studio project/solution (#12625)
|
10 bulan lalu |
|
backend.hpp
|
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)
|
8 bulan lalu |
|
binbcast.cpp
|
8d66005763
SYCL: Refactor and enable FP16 in binary broadcast OPs (#12975)
|
9 bulan lalu |
|
binbcast.hpp
|
8d66005763
SYCL: Refactor and enable FP16 in binary broadcast OPs (#12975)
|
9 bulan lalu |
|
common.cpp
|
6c02a032fa
SYCL: Remove misleading ggml_sycl_op_flatten function (#12387)
|
10 bulan lalu |
|
common.hpp
|
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)
|
8 bulan lalu |
|
concat.cpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
concat.hpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
conv.cpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
conv.hpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
convert.cpp
|
8733e0cf6e
sycl: addressing non-contiguous src1 mul_mats (nc and batched) (#13343)
|
9 bulan lalu |
|
convert.hpp
|
8733e0cf6e
sycl: addressing non-contiguous src1 mul_mats (nc and batched) (#13343)
|
9 bulan lalu |
|
cpy.cpp
|
ece9745bb8
SYCL: Move CPY kernels to a separate file and add few missing kernels (#12133)
|
11 bulan lalu |
|
cpy.hpp
|
ece9745bb8
SYCL: Move CPY kernels to a separate file and add few missing kernels (#12133)
|
11 bulan lalu |
|
dequantize.hpp
|
08d5986290
[SYCL] Optimize mul_mat for Q4_0 on Intel GPU (#12035)
|
11 bulan lalu |
|
dmmv.cpp
|
a53f7f7b88
fixed compilation warnings in ggml-sycl (#12424)
|
10 bulan lalu |
|
dmmv.hpp
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 tahun lalu |
|
element_wise.cpp
|
a4c340f974
SYCL: Add all missing unary kernels (#13074)
|
9 bulan lalu |
|
element_wise.hpp
|
a4c340f974
SYCL: Add all missing unary kernels (#13074)
|
9 bulan lalu |
|
gemm.hpp
|
9ffcc9e374
sycl: cleanup oneDNN related code (#12097)
|
10 bulan lalu |
|
getrows.cpp
|
6c02a032fa
SYCL: Remove misleading ggml_sycl_op_flatten function (#12387)
|
10 bulan lalu |
|
getrows.hpp
|
6c02a032fa
SYCL: Remove misleading ggml_sycl_op_flatten function (#12387)
|
10 bulan lalu |
|
ggml-sycl.cpp
|
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)
|
8 bulan lalu |
|
gla.cpp
|
f446c2cf6a
SYCL: Add gated linear attention kernel (#11175)
|
1 tahun lalu |
|
gla.hpp
|
f446c2cf6a
SYCL: Add gated linear attention kernel (#11175)
|
1 tahun lalu |
|
im2col.cpp
|
75afa0ae31
SYCL: Fix im2col (#12910)
|
9 bulan lalu |
|
im2col.hpp
|
6c02a032fa
SYCL: Remove misleading ggml_sycl_op_flatten function (#12387)
|
10 bulan lalu |
|
mmq.cpp
|
a53f7f7b88
fixed compilation warnings in ggml-sycl (#12424)
|
10 bulan lalu |
|
mmq.hpp
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 tahun lalu |
|
mmvq.cpp
|
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)
|
8 bulan lalu |
|
mmvq.hpp
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 tahun lalu |
|
norm.cpp
|
8bbf26083d
SYCL: switch to SYCL namespace (#12674)
|
10 bulan lalu |
|
norm.hpp
|
6c02a032fa
SYCL: Remove misleading ggml_sycl_op_flatten function (#12387)
|
10 bulan lalu |
|
outprod.cpp
|
8293970542
SYCL: Rename oneMKL to oneMath (#12192)
|
10 bulan lalu |
|
outprod.hpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
presets.hpp
|
3bcd40b3c5
Optimize RWKV6 Operator Naming and Implement Multi-core CPU/ SYCL Acceleration (#10133)
|
1 tahun lalu |
|
quants.hpp
|
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)
|
8 bulan lalu |
|
rope.cpp
|
5368ddda7a
SYCL: Add non-contiguous support in ROPE (#12993)
|
9 bulan lalu |
|
rope.hpp
|
5368ddda7a
SYCL: Add non-contiguous support in ROPE (#12993)
|
9 bulan lalu |
|
softmax.cpp
|
a53f7f7b88
fixed compilation warnings in ggml-sycl (#12424)
|
10 bulan lalu |
|
softmax.hpp
|
6e84b0ab8e
SYCL : SOFTMAX F16 mask support and other fixes (#11261)
|
1 tahun lalu |
|
sycl_hw.cpp
|
08d5986290
[SYCL] Optimize mul_mat for Q4_0 on Intel GPU (#12035)
|
11 bulan lalu |
|
sycl_hw.hpp
|
08d5986290
[SYCL] Optimize mul_mat for Q4_0 on Intel GPU (#12035)
|
11 bulan lalu |
|
tsembd.cpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
tsembd.hpp
|
c6860cc734
SYCL: Refactor ggml_sycl_compute_forward (#11121)
|
1 tahun lalu |
|
vecdotq.hpp
|
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)
|
8 bulan lalu |
|
wkv.cpp
|
7dfad387e3
llama: Add support for RWKV v7 architecture (#12412)
|
10 bulan lalu |
|
wkv.hpp
|
7dfad387e3
llama: Add support for RWKV v7 architecture (#12412)
|
10 bulan lalu |
Alberto Cabrera Pérez
17512a94d6
sycl : implementation of reordered Q4_0 MMVQ for Intel GPUs (#12858)