 Howard Su
|
b922bc351b
llama : remove shards weight file support (#2000)
|
2 лет назад |
 Johannes Gäßler
|
7f9753fa12
CUDA GPU acceleration for LoRAs + f16 models (#1970)
|
2 лет назад |
 ningshanwutuobang
|
cfa0750bc9
llama : support input embeddings directly (#1910)
|
2 лет назад |
 Erik Scholz
|
9d23589d63
fix pthreads setaffinity usage on android (#2020)
|
2 лет назад |
|
Howard Su
|
0be54f75a6
baby-llama : fix build after ggml_rope change (#2016)
|
2 лет назад |
 Georgi Gerganov
|
181e8d9755
llama : fix rope usage after ChatGLM change
|
2 лет назад |
|
Georgi Gerganov
|
d9779021bd
ggml : add support for ChatGLM RoPE
|
2 лет назад |
 Roman Parykin
|
d38e451578
readme : add Scala 3 bindings repo (#2010)
|
2 лет назад |
 David Yang
|
eaa6ca5a61
ggml : increase max tensor name + clean up compiler warnings in train-text (#1988)
|
2 лет назад |
 Gustavo Rocha Dias
|
aa777abbb7
readme : LD_LIBRARY_PATH complement for some Android devices when building with CLBlast inside Termux (#2007)
|
2 лет назад |
|
Georgi Gerganov
|
c824d2e368
ggml : avoid conv 2d kernel round up
|
2 лет назад |
 zrm
|
b853d45601
ggml : add NUMA support (#1556)
|
2 лет назад |
|
Georgi Gerganov
|
9225baef71
k-quants : fix indentation
|
2 лет назад |
 katsu560
|
a84ab1da8d
tests : fix quantize perf (#1990)
|
2 лет назад |
|
katsu560
|
5743ca8092
k-quants : add AVX support to dot functions (#1916)
|
2 лет назад |
|
Georgi Gerganov
|
412c60e473
readme : add link to new k-quants for visibility
|
2 лет назад |
 Kawrakow
|
6769e944c7
k-quants : support for super-block size of 64 (#2001)
|
2 лет назад |
|
Howard Su
|
cbebf61ca7
Fix assert when free invalid cuda pointer (#2005)
|
2 лет назад |
|
Georgi Gerganov
|
447ccbe8c3
readme : add new roadmap + manifesto
|
2 лет назад |
|
Georgi Gerganov
|
bd34cdde38
ggml : sync latest ggml (custom operators)
|
2 лет назад |
 anon998
|
c2a08f87b8
fix server sampling: top k sampler first (#1977)
|
2 лет назад |
|
Georgi Gerganov
|
66a2555ba6
readme : add Azure CI discussion link
|
2 лет назад |
 sjinzh
|
e65ca7e14a
zig : upgrade build system support (#1981)
|
2 лет назад |
 Robyn
|
5ec8dd5a3c
#1869 Fix null reference errors when training from scratch with CUDA (#1907)
|
2 лет назад |
|
Georgi Gerganov
|
65bdd52a86
tests : sync test-grad0 from ggml
|
2 лет назад |
 Rowan Hart
|
fdd1860911
flake : fix ggml-metal.metal path and run nixfmt (#1974)
|
2 лет назад |
 AN Long
|
c943d823c1
convert : fix invalid params in write_vocab_only (#1975)
|
2 лет назад |
 slaren
|
f2c754e1c3
ggml : improve ggml_graph_dump_dot, add ggml_format_name (#1978)
|
2 лет назад |
|
Georgi Gerganov
|
11da1a85cd
readme : fix whitespaces
|
2 лет назад |
 Alberto
|
235b610d65
readme : fixed termux instructions (#1973)
|
2 лет назад |
