 Georgi Gerganov
|
2f9ec7e271
cuda : improve text-generation and batched decoding performance (#3776)
|
il y a 2 ans |
|
Georgi Gerganov
|
2b4ea35e56
cuda : add batched cuBLAS GEMM for faster attention (#3749)
|
il y a 2 ans |
|
Georgi Gerganov
|
d28e572c02
cmake : fix add_compile_options on macOS
|
il y a 2 ans |
|
Georgi Gerganov
|
db3abcc114
sync : ggml (ggml-backend) (#3548)
|
il y a 2 ans |
 Eve
|
017efe899d
cmake : make LLAMA_NATIVE flag actually use the instructions supported by the processor (#3273)
|
il y a 2 ans |
 cebtenzzre
|
e78f0b0d05
cmake : increase minimum version for add_link_options (#3444)
|
il y a 2 ans |
|
cebtenzzre
|
9476b01226
cmake : make CUDA flags more similar to the Makefile (#3420)
|
il y a 2 ans |
 bandoti
|
095231dfd3
cmake : fix transient definitions in find pkg (#3411)
|
il y a 2 ans |
|
Cebtenzzre
|
bc39553c90
build : enable more non-default compiler warnings (#3200)
|
il y a 2 ans |
 Jag Chadha
|
527e57cfd8
build : add ACCELERATE_NEW_LAPACK to fix warning on macOS Sonoma (#3342)
|
il y a 2 ans |
 DAN™
|
99115f3fa6
cmake : fix build-info.h on MSVC (#3309)
|
il y a 2 ans |
 Johannes Gäßler
|
111163e246
CUDA: enable peer access between devices (#2470)
|
il y a 2 ans |
|
Cebtenzzre
|
3aefaab9e5
check C++ code with -Wmissing-declarations (#3184)
|
il y a 2 ans |
 Engininja2
|
7e50d34be6
cmake : fix building shared libs for clang (rocm) on windows (#3176)
|
il y a 2 ans |
 Andrei
|
76164fe2e6
cmake : fix llama.h location when built outside of root directory (#3179)
|
il y a 2 ans |
|
Andrei
|
769266a543
cmake : compile ggml-rocm with -fpic when building shared library (#3158)
|
il y a 2 ans |
|
bandoti
|
990a5e226a
cmake : add relocatable Llama package (#2960)
|
il y a 2 ans |
 Tristan Ross
|
1b6c650d16
cmake : add a compiler flag check for FP16 format (#3086)
|
il y a 2 ans |
|
Johannes Gäßler
|
0a5eebb45d
CUDA: mul_mat_q RDNA2 tunings (#2910)
|
il y a 2 ans |
 Eric Sommerlade
|
b52b29ab9d
arm64 support for windows (#3007)
|
il y a 2 ans |
 Jhen-Jie Hong
|
1b0d09259e
cmake : support build for iOS/tvOS (#3116)
|
il y a 2 ans |
|
Georgi Gerganov
|
b3e9852e47
sync : ggml (CUDA GLM RoPE + POSIX) (#3082)
|
il y a 2 ans |
 Przemysław Pawełczyk
|
cb6c44c5e0
build : do not use _GNU_SOURCE gratuitously (#2035)
|
il y a 2 ans |
 Kunshang Ji
|
7f412dab9c
enable CPU HBM (#2603)
|
il y a 2 ans |
|
Cebtenzzre
|
00d62adb79
fix some warnings from gcc and clang-tidy (#3038)
|
il y a 2 ans |
|
Cebtenzzre
|
9912b9efc8
build : add LLAMA_METAL_NDEBUG flag (#3033)
|
il y a 2 ans |
|
Georgi Gerganov
|
e36ecdccc8
build : on Mac OS enable Metal by default (#2901)
|
il y a 2 ans |
|
Cebtenzzre
|
ef15649972
build : fix most gcc and clang warnings (#2861)
|
il y a 2 ans |
|
Cebtenzzre
|
849408957c
tests : add a C compliance test (#2848)
|
il y a 2 ans |
|
Georgi Gerganov
|
3a007648f2
metal : add option to disable debug logs (close #2764)
|
il y a 2 ans |
