ngxson/llama.cpp

Pull Requests Commits

clean up json_value & server_log

ngxson committed 1 year ago

cb46be70

compare-llama-bench.py: add missing basicConfig (#7138)

mofosyne committed 1 year ago

Verified acdce3cd

ggml : introduce bfloat16 support (#6412)

jart committed 1 year ago

Verified 38554160

metal : fix unused warning

ggerganov committed 1 year ago

Verified c0e6fbf8

Further tidy on Android instructions README.md (#7077)

Jeximo committed 1 year ago

Verified c780e753

Fixed save_imatrix to match old behaviour for MoE (#7099)

jukofyork committed 1 year ago

Verified 48b2f9c1

server: fix incorrectly reported token probabilities (#7125)

JohannesGaessler committed 1 year ago

Verified af0a5b61

Fix OLMo HF to GGUF conversion (#6910)

nopperl committed 1 year ago

Verified b6aa6702

server : update readme with undocumented options (#7013)

K-Mistele committed 1 year ago

Verified 260b7c65

readme : update hot topics

ggerganov committed 1 year ago

Verified 53d6c52e

main : update log text (EOS to EOG) (#7104)

RhinoDevel committed 1 year ago

Verified 3af34c1d

docs: fix typos (#7124)

omahs committed 1 year ago

Verified 04976db7

ci : add GG_BUILD_EXTRA_TESTS_0 env (#7098)

ggerganov committed 1 year ago

Verified 947d3ad2

Add an option to build without CUDA VMM (#7067)

WilliamTambellini committed 1 year ago

Verified 858f6b73

flake.lock: Update (#7079)

ggerganov committed 1 year ago

Verified b3a995b4

minor : fix trailing whitespace

ggerganov committed 1 year ago

Verified bcdee0da

Adding support for the --numa argument for llama-bench. (#7080)

kunnis committed 1 year ago

Verified 628b2991

Disable benchmark on forked repo (#7034)

CISC committed 1 year ago

Verified 8f8acc86

readme : add note that LLaMA 3 is not supported with convert.py (#7065)

lyledean1 committed 1 year ago

Verified ca363260

command-r : add BPE pre-tokenization (#7063)

dranger003 committed 1 year ago

Verified 889bdd76

py : logging and flake8 suppression refactoring (#7081)

mofosyne committed 1 year ago

Verified 6fbd4322

gguf-split: add --no-tensor-first-split (#7072)

ngxson committed 1 year ago

Verified 84250014

Tidy Android Instructions README.md (#7016)

Jeximo committed 1 year ago

Verified cf768b7e

Fix Linux /sys cpu path to guess number of cores (#7064)

viric committed 1 year ago

Verified fcd84a0f

If first token generated from the server is the stop word the server will crash (#7038)

maor-ps committed 1 year ago

Verified 03fb8a00

tests : add test-tokenizer-0.sh + fix some tokenizers (#7036)

ggerganov committed 1 year ago

Verified 92139b90

convert.py : add python logging instead of print() (#6511)

mofosyne committed 1 year ago

Verified a2ac89d6

llama : rename ctx to user_data in progress_callback (#7045)

danbev committed 1 year ago

Verified 433def28

Remove .attention from skipped tensors to match more accurately (#7051)

bartowski1182 committed 1 year ago

Verified 60325fa5

chore: fix typo in llama.cpp (#7032)

alwqx committed 1 year ago

Verified 6ecf3189

Older