صفحهٔ اصلی گشت‌و‌گذار راهنما
ورود
cturan
/
llama.cpp
mirrorاز https://github.com/cturan/llama.cpp
1
0
انشعاب 0
پرونده‌ها مشکلات 0 ویکی
درخت: b49a13dd2f
شاخه‌ها تگ‌ها
llama.cpp
/
ggml-cuda
Johannes Gäßler 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
..
acc.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
acc.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
arange.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
arange.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
argsort.cu 08a0c02060 ggml : mul_mat_id use the same tensor for all the experts (#6387) 1 سال پیش
argsort.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
binbcast.cu 0d56246f4b ggml : group all experts in a single ggml_mul_mat_id (#6505) 1 سال پیش
binbcast.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
clamp.cu bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
clamp.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
common.cuh dc685be466 CUDA: add FP32 FlashAttention vector kernel (#7188) 1 سال پیش
concat.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
concat.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
convert.cu bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
convert.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 سال پیش
cpy.cu bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
cpy.cuh bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
dequantize.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 سال پیش
diagmask.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
diagmask.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
dmmv.cu 5dc9dd7152 llama : add Command R Plus support (#6491) 1 سال پیش
dmmv.cuh d48ccf3ad4 sync : ggml (#6351) 1 سال پیش
fattn-common.cuh dc685be466 CUDA: add FP32 FlashAttention vector kernel (#7188) 1 سال پیش
fattn-tile-f16.cu 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn-tile-f16.cuh 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn-tile-f32.cu 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn-tile-f32.cuh 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn-vec-f16.cu 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn-vec-f16.cuh dc685be466 CUDA: add FP32 FlashAttention vector kernel (#7188) 1 سال پیش
fattn-vec-f32.cu 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn-vec-f32.cuh dc685be466 CUDA: add FP32 FlashAttention vector kernel (#7188) 1 سال پیش
fattn.cu 0fc1e820a9 CUDA: faster large batch FA without tensor cores (#7314) 1 سال پیش
fattn.cuh 9c67c2773d ggml : add Flash Attention (#5021) 1 سال پیش
getrows.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
getrows.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
im2col.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
im2col.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
mmq.cu bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
mmq.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
mmvq.cu bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
mmvq.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
norm.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
norm.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
pad.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
pad.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
pool2d.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
pool2d.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
quantize.cu 5dc9dd7152 llama : add Command R Plus support (#6491) 1 سال پیش
quantize.cuh 5dc9dd7152 llama : add Command R Plus support (#6491) 1 سال پیش
rope.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
rope.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
scale.cu bc4bba364f Introduction of CUDA Graphs to LLama.cpp (#6766) 1 سال پیش
scale.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
softmax.cu 9cb317f77e ggml : full ALiBi support (#7192) 1 سال پیش
softmax.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
sumrows.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
sumrows.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
tsembd.cu ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
tsembd.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
unary.cu f5ef34e428 feat: implemented sigmoid function (ggml/806) 1 سال پیش
unary.cuh f5ef34e428 feat: implemented sigmoid function (ggml/806) 1 سال پیش
upscale.cu 48aa8fd1f2 ggml : add `ggml_upscale_ext` (ggml/814) 1 سال پیش
upscale.cuh ae1f211ce2 cuda : refactor into multiple files (#6269) 1 سال پیش
vecdotq.cuh 55c1b2a3bb IQ1_M: 1.75 bpw quantization (#6302) 1 سال پیش