Skip to content
GitLab
Explore
Sign in
Tags
Tags give the ability to mark specific points in history as being important
b1269
51a7cf5c
·
examples : fix RoPE defaults to match PR #3240 (#3315)
·
Sep 23, 2023
b1266
36b904e2
·
ggml-opencl.cpp: Make private functions static (#3300)
·
Sep 21, 2023
b1263
8185710a
·
CUDA: use only 1 thread if fully offloaded (#2915)
·
Sep 21, 2023
b1261
a5661d7e
·
llama : allow gguf RoPE keys to be overridden with defaults (#3240)
·
Sep 20, 2023
b1260
65c2c1c5
·
benchmark-matmult : do not use integer abs() on a float (#3277)
·
Sep 20, 2023
b1258
a40f2b65
·
CI: FreeBSD fix (#3258)
·
Sep 20, 2023
b1257
d119c04c
·
examples : fix benchmark-matmult (#1554)
·
Sep 20, 2023
b1256
8781013e
·
make : restore build-info.h dependency for several targets (#3205)
·
Sep 18, 2023
b1255
7ddf1855
·
ci : switch cudatoolkit install on windows to networked (#3236)
·
Sep 18, 2023
b1254
ee66942d
·
CUDA: fix peer access logic (#3231)
·
Sep 17, 2023
b1253
111163e2
·
CUDA: enable peer access between devices (#2470)
·
Sep 17, 2023
b1252
8b428c9b
·
llama.cpp : show model size and BPW on load (#3223)
·
Sep 17, 2023
b1251
578d8c8f
·
CUDA: fix scratch malloced on non-main device (#3220)
·
Sep 17, 2023
b1250
b541b4f0
·
Enable BUILD_SHARED_LIBS=ON on all Windows builds (#3215)
·
Sep 16, 2023
master-0631ea3
0631ea36
·
Don't crash on available devices if we can't even create an instance.
·
Sep 16, 2023
b1249
5dbc2b32
·
Enable build with CUDA 11.0 (make) (#3132)
·
Sep 16, 2023
b1248
b08e75ba
·
Fixing the last deviations from sentencepiece indicated by test-tokenizer-1 (#3170)
·
Sep 16, 2023
b1247
e6616cf0
·
examples : add compiler version and target to build info (#2998)
·
Sep 15, 2023
b1246
3aefaab9
·
check C++ code with -Wmissing-declarations (#3184)
·
Sep 15, 2023
b1245
69eb67e2
·
fix build numbers by setting fetch-depth=0 (#3197)
·
Sep 15, 2023
Prev
1
…
55
56
57
58
59
60
61
62
63
…
99
Next