| .. |
|
template-instances
|
8e558309dc
CUDA: MMQ support for iq4_nl, iq4_xs (#8278)
|
1 年之前 |
|
acc.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
acc.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
arange.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
arange.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
argsort.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
argsort.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
binbcast.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
binbcast.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
clamp.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
clamp.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
common.cuh
|
d23287f122
Define and optimize RDNA1 (#8085)
|
1 年之前 |
|
concat.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
concat.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
conv-transpose-1d.cu
|
fde13b3bb9
feat: cuda implementation for `ggml_conv_transpose_1d` (ggml/854)
|
1 年之前 |
|
conv-transpose-1d.cuh
|
fde13b3bb9
feat: cuda implementation for `ggml_conv_transpose_1d` (ggml/854)
|
1 年之前 |
|
convert.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
convert.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
cpy.cu
|
07a3fc0608
Removes multiple newlines at the end of files that is breaking the editorconfig step of CI. (#8258)
|
1 年之前 |
|
cpy.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
dequantize.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
diagmask.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
diagmask.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
dmmv.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
dmmv.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-common.cuh
|
8e558309dc
CUDA: MMQ support for iq4_nl, iq4_xs (#8278)
|
1 年之前 |
|
fattn-tile-f16.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-tile-f16.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-tile-f32.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-tile-f32.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-vec-f16.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-vec-f32.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn-wmma-f16.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
fattn.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
getrows.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
getrows.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
im2col.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
im2col.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
mma.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
mmq.cu
|
8e558309dc
CUDA: MMQ support for iq4_nl, iq4_xs (#8278)
|
1 年之前 |
|
mmq.cuh
|
8e558309dc
CUDA: MMQ support for iq4_nl, iq4_xs (#8278)
|
1 年之前 |
|
mmvq.cu
|
cb5fad4c6c
CUDA: refactor and optimize IQ MMVQ (#8215)
|
1 年之前 |
|
mmvq.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
norm.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
norm.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
pad.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
pad.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
pool2d.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
pool2d.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
quantize.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
quantize.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
rope.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
rope.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
scale.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
scale.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
softmax.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
softmax.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
sumrows.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
sumrows.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
tsembd.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
tsembd.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
unary.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
unary.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
upscale.cu
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
upscale.cuh
|
f3f65429c4
llama : reorganize source code + improve CMake (#8006)
|
1 年之前 |
|
vecdotq.cuh
|
8e558309dc
CUDA: MMQ support for iq4_nl, iq4_xs (#8278)
|
1 年之前 |
John Balis