Skip to content
GitLab
Explore
Sign in
Tags
Tags give the ability to mark specific points in history as being important
master-dfd9fce
dfd9fce6
·
ggml : fix restrict usage
·
Jul 06, 2023
master-36680f6
36680f6e
·
convert : update for baichuan (#2081)
·
Jul 06, 2023
master-31cfbb1
31cfbb10
·
Expose generation timings from server & update completions.js (#2116)
·
Jul 05, 2023
master-ec326d3
ec326d35
·
ggml : fix bug introduced in #1237
·
Jul 05, 2023
master-1b6efea
1b6efeab
·
tests : fix test-grad0
·
Jul 05, 2023
master-1b107b8
1b107b85
·
ggml : generalize `quantize_fns` for simpler FP16 handling (#1237)
·
Jul 05, 2023
master-924dd22
924dd22f
·
Quantized dot products for CUDA mul mat vec (#2067)
·
Jul 05, 2023
master-051c70d
051c70dc
·
llama: Don't double count the sampling time (#2107)
·
Jul 05, 2023
master-9e4475f
9e4475f5
·
Fixed OpenCL offloading prints (#2082)
·
Jul 05, 2023
master-7f0e9a7
7f0e9a77
·
embd-input: Fix input embedding example unsigned int seed (#2105)
·
Jul 05, 2023
master-ed9a54e
ed9a54e5
·
ggml : sync latest (new ops, macros, refactoring) (#2106)
·
Jul 04, 2023
master-f257fd2
f257fd25
·
Add an API example using server.cpp similar to OAI. (#2009)
·
Jul 04, 2023
master-7ee76e4
7ee76e45
·
Simple webchat for server (#1998)
·
Jul 04, 2023
master-acc111c
acc111ca
·
Allow old Make to build server. (#2098)
·
Jul 04, 2023
master-23c7c6f
23c7c6fc
·
Update Makefile: clean simple (#2097)
·
Jul 04, 2023
master-698efad
698efad5
·
CI: make the brew update temporarily optional. (#2092)
·
Jul 04, 2023
master-d7d2e6a
d7d2e6a0
·
server: add option to output probabilities for completion (#1962)
·
Jul 03, 2023
master-46088f7
46088f72
·
ggml : fix build with OpenBLAS (close #2066)
·
Jul 02, 2023
master-0bc2cdf
0bc2cdfc
·
Better CUDA synchronization logic (#2057)
·
Jul 01, 2023
master-2f8cd97
2f8cd979
·
metal : release buffers when freeing metal context (#2062)
·
Jul 01, 2023
Prev
1
…
71
72
73
74
75
76
77
78
79
…
99
Next