 jukofyork
|
48b2f9c1fc
Fixed save_imatrix to match old behaviour for MoE (#7099)
|
il y a 1 an |
 Johannes Gäßler
|
af0a5b6163
server: fix incorrectly reported token probabilities (#7125)
|
il y a 1 an |
 nopperl
|
b6aa670203
Fix OLMo HF to GGUF conversion (#6910)
|
il y a 1 an |
 Kyle Mistele
|
260b7c6529
server : update readme with undocumented options (#7013)
|
il y a 1 an |
 Georgi Gerganov
|
53d6c52e22
readme : update hot topics
|
il y a 1 an |
 RhinoDevel
|
3af34c1d1b
main : update log text (EOS to EOG) (#7104)
|
il y a 1 an |
 omahs
|
04976db7a8
docs: fix typos (#7124)
|
il y a 1 an |
|
Georgi Gerganov
|
947d3ad27d
ci : add GG_BUILD_EXTRA_TESTS_0 env (#7098)
|
il y a 1 an |
 William Tambellini
|
858f6b73f6
Add an option to build without CUDA VMM (#7067)
|
il y a 1 an |
|
Georgi Gerganov
|
b3a995b416
flake.lock: Update (#7079)
|
il y a 1 an |
|
Georgi Gerganov
|
bcdee0daa7
minor : fix trailing whitespace
|
il y a 1 an |
 kunnis
|
628b299106
Adding support for the --numa argument for llama-bench. (#7080)
|
il y a 1 an |
 Sigbjørn Skjæret
|
8f8acc8683
Disable benchmark on forked repo (#7034)
|
il y a 1 an |
 Lyle Dean
|
ca36326020
readme : add note that LLaMA 3 is not supported with convert.py (#7065)
|
il y a 1 an |
 DAN™
|
889bdd7686
command-r : add BPE pre-tokenization (#7063)
|
il y a 1 an |
 Brian
|
6fbd432211
py : logging and flake8 suppression refactoring (#7081)
|
il y a 1 an |
 Xuan Son Nguyen
|
842500144e
gguf-split: add --no-tensor-first-split (#7072)
|
il y a 1 an |
 Jeximo
|
cf768b7e71
Tidy Android Instructions README.md (#7016)
|
il y a 1 an |
 viric
|
fcd84a0f5a
Fix Linux /sys cpu path to guess number of cores (#7064)
|
il y a 1 an |
 maor-ps
|
03fb8a002d
If first token generated from the server is the stop word the server will crash (#7038)
|
il y a 1 an |
|
Georgi Gerganov
|
92139b90af
tests : add test-tokenizer-0.sh + fix some tokenizers (#7036)
|
il y a 1 an |
|
Brian
|
a2ac89d6ef
convert.py : add python logging instead of print() (#6511)
|
il y a 1 an |
 Daniel Bevenius
|
433def286e
llama : rename ctx to user_data in progress_callback (#7045)
|
il y a 1 an |
 Bartowski
|
60325fa56f
Remove .attention from skipped tensors to match more accurately (#7051)
|
il y a 1 an |
 alwqx
|
6ecf3189e0
chore: fix typo in llama.cpp (#7032)
|
il y a 1 an |
 Andrew Downing
|
b0d943de17
Update LOG_IMPL and LOG_TEE_IMPL (#7029)
|
il y a 1 an |
 l3utterfly
|
8d608a81b7
main : fix off by one error for context shift (#6921)
|
il y a 1 an |
|
Johannes Gäßler
|
3ea0d36000
Server: add tests for batch size, different seeds (#6950)
|
il y a 1 an |
|
Johannes Gäßler
|
1613ef8d8e
CUDA: CUDART < 11.7 workaround for __hmax, __hmax2 (#7019)
|
il y a 1 an |
 slaren
|
c4ec9c0d3d
ci : exempt confirmed bugs from being tagged as stale (#7014)
|
il y a 1 an |
