Admin message
为了安全，强烈建议开启2FA双因子认证：User Settings -> Account -> Enable two-factor authentication！！！
Tags
Tags give the ability to mark specific points in history as being important
b4016
42cadc74
·
server : fix slot selection by lru (#10126)
·
Nov 02, 2024
b4015
45950415
·
server : fix endpoint checks (#10135)
·
Nov 02, 2024
b4014
1926d6e3
·
llama : adjust default context size + print warnings (#10136)
·
Nov 02, 2024
b4013
b634f8a2
·
simple-chat : only add bos on first prompt (#10129)
·
Nov 02, 2024
b4011
a6744e43
·
llama : add simple-chat example (#10124)
·
Nov 01, 2024
b4010
e991e312
·
llama : use smart pointers for ggml resources (#10117)
·
Nov 01, 2024
b4009
418f5eef
·
vulkan : improve ggml_vk_create_buffer error handling (#9898)
·
Nov 01, 2024
b4007
d865d147
·
server : fix smart selection of available slot (#10120)
·
Nov 01, 2024
b4006
1804adb0
·
ggml : remove ggml_scratch (#10121)
·
Nov 01, 2024
b4005
815fe72a
·
sync : ggml
·
Nov 01, 2024
b4003
e597e507
·
build: fix build error in Windows env with OneAPI setup (#10107)
·
Nov 01, 2024
b4002
85679d37
·
llama : improve output buffer type selection (#10098)
·
Nov 01, 2024
b4001
1e9f9499
·
quantize : fix --keep-split (#10114)
·
Nov 01, 2024
b4000
c02e5ab2
·
llama : fix buffer checks for mamba and rwk (#10111)
·
Oct 31, 2024
b3999
ab3d71f9
·
loader: refactor tensor weights storage (#9935)
·
Oct 31, 2024
b3998
0a683e80
·
server : include scheme when printing URL (#10106)
·
Oct 31, 2024
b3997
dea5e860
·
ggml : check tensor name lengths in gguf files (#10100)
·
Oct 31, 2024
b3996
1329c0a7
·
kompute: add mul_mat_q4_k shader (#10097)
·
Oct 31, 2024
b3995
61408e7f
·
kompute: add backend registry / device interfaces (#10045)
·
Oct 30, 2024
b3994
b9e02e81
·
ggml : fix memory leaks when loading invalid gguf files (#10094)
·
Oct 30, 2024
1
…
43
44
45
46
47
48
49
50
51
…
178