 Kawrakow
|
ca82cf7bac
metal : more optimizations (#2959)
|
2 anni fa |
 kchro3
|
6a31a3bd98
swift : add support for k-quants (#2983)
|
2 anni fa |
 Kerfuffle
|
cff7b0bf07
convert.py : BPE fixes (#2938)
|
2 anni fa |
 Ido S
|
340af42f09
docs : add `catai` to `README.md` (#2967)
|
2 anni fa |
 momonga
|
c42f0ec6b3
examples : fix gpt-neox (#2943)
|
2 anni fa |
|
kchro3
|
2753415afd
swift : add missing c file to Package.swift (#2978)
|
2 anni fa |
 Cebtenzzre
|
bc054af97a
make : support overriding CFLAGS/CXXFLAGS/CPPFLAGS/LDFLAGS (#2886)
|
2 anni fa |
|
Kerfuffle
|
3358c381f6
logging: Fix creating empty file even when disabled (#2966)
|
2 anni fa |
 bandoti
|
52315a4216
readme : update clblast instructions (#2903)
|
2 anni fa |
 Karsten Weiss
|
8b56b4f2c3
metal : show all Metal device instances in the system (#2952)
|
2 anni fa |
 Jhen-Jie Hong
|
21f3d1be86
k-quants : fix build on armv7 (android only) (#2920)
|
2 anni fa |
|
Jhen-Jie Hong
|
571083f508
server : avoid aniprompt in probabilities of final response (#2849)
|
2 anni fa |
 Engininja2
|
f04d002844
cuda : vsubss4 for older versions of ROCm/clang (#2942)
|
2 anni fa |
 ZHAOKAI WANG
|
69fdbb9abc
readme : quick start command fix (#2908)
|
2 anni fa |
|
Kerfuffle
|
5d6f19f16b
Allow quantize to only copy tensors, some other improvements (#2931)
|
2 anni fa |
 Georgi Gerganov
|
0d58936686
llama2c : rename function
|
2 anni fa |
|
Cebtenzzre
|
6c9c23429b
make : use unaligned vector moves on MinGW (#2945)
|
2 anni fa |
 m3ndax
|
ee8654bcd0
minor : add const qualifiers (#2853)
|
2 anni fa |
 Konstantin Herud
|
49bb9cbe0f
docs : add java-llama.cpp to README.md (#2935)
|
2 anni fa |
|
Cebtenzzre
|
ef15649972
build : fix most gcc and clang warnings (#2861)
|
2 anni fa |
 Ben Siraphob
|
d8d6977f48
examples : add C grammar (#2357)
|
2 anni fa |
 Tameem
|
5aec2cfaac
ggml : add RISC-V vector intrinsics support (#2929)
|
2 anni fa |
|
Georgi Gerganov
|
13268c5331
metal : slight speed-up for add and mul kernels (#2917)
|
2 anni fa |
 staviq
|
4dcd47d71d
logs : fix mingw-like builds (fixes #2898) (#2911)
|
2 anni fa |
|
Cebtenzzre
|
18705a30ef
llama2c : fix segfault and alloc-dealloc-mismatch (#2913)
|
2 anni fa |
|
Kawrakow
|
e8d9158925
metal: somewhat faster f16 x f32 matrix multiply kernel (#2951)
|
2 anni fa |
|
Cebtenzzre
|
bce1fef328
convert : fix another python 3.8 issue (#2949)
|
2 anni fa |
 slaren
|
528134dd02
remove convert-llama-7b-pth-to-gguf.py and convert-llama-hf-to-gguf.py (#2906)
|
2 anni fa |
|
Kerfuffle
|
aeefac4ff7
scripts: Use local gguf package when running from repo (#2927)
|
2 anni fa |
 DannyDaemonic
|
e8422de39e
@vxiiduu's fix for PrefetchVirtualMemory (#2930)
|
2 anni fa |
