Skip to content

LLM inference in C/C++

C++ ggml

Commits on master

vulkan: MOE aware mat_mul_id tile selection (#29182)
Ankit Khandelwal committed 1 month ago
94a0ae3
a3f84fa
284153e
ggml : require input tensors to be GGML_OP_NONE (#29647)
Georgi Gerganov committed 1 month ago
b5cf8ce
e904318
d280808
common : use fs::path for config dir (#29649)
Adrien Gallouët committed 1 month ago
00af635
common : add fs_write_atomic() (#29642)
Adrien Gallouët committed 1 month ago
31385c9
8019dc5
ggml-zdnn: fix 0-row tensor crash (#29636)
Aaron Teo committed 1 month ago
86ea01d
ggml : speed up model loading (#29598)
Adrien Gallouët committed 1 month ago
c13e04e
c8cda8b
llama : fix init in several tools/examples (#29632)
Georgi Gerganov committed 1 month ago
6d78fb0
metal: FWHT perf optimizations (#29602)
bri-prism committed 1 month ago
18bbc46
common : use fs::path for cache dirs (#29595)
Adrien Gallouët committed 1 month ago
76a5bc8
46e17a6
fc07d78
vulkan: include functional header (#29597)
Linux User committed 1 month ago
66e665c
57b557c
14ebbd5
f1ea206