ggml-org/llama.cpp

Pull Requests Commits

Merge branch 'master' of https://github.com/ggerganov/llama.cpp into ceb/fix-cuda-warning-flags

cebtenzzre committed 1 year ago

c8554b80

get_flags.mk -> get-flags.mk

cebtenzzre committed 1 year ago

d870a9fd

common : add `--version` option to show build info in CLI (#4433)

yusiwen committed 1 year ago

Verified 9fb13f95

readme : update hot topics

ggerganov committed 1 year ago

Verified 113f9942

llama : add Mixtral support (#4406)

slaren committed 1 year ago

Verified 799a1cb1

cmake : fix improper joining in generator expression

cebtenzzre committed 1 year ago

cacac251

cmake : make CUDA warning stuff properly conditional

cebtenzzre committed 1 year ago

cdf3cc3c

cmake : capitalize variables

cebtenzzre committed 1 year ago

e30a8ad1

cmake : fix incorrect variable reference

cebtenzzre committed 1 year ago

b5b2cdff

server : tweak default sampling parameters (#4367)

kalomaze committed 1 year ago

Verified fecac456

english : use `typos` to fix comments and logs (#4354)

richardkiss committed 1 year ago

Verified 9494d7c4

build : target Windows 8 for standard mingw-w64 (#4405)

cebtenzzre committed 1 year ago

Verified 6138963f

llama : document logits_all deprecation (#4418)

crasm committed 1 year ago

Verified 6391817c

server : fix local model name in server (#4420)

epicfilemcnulty committed 1 year ago

Verified d9d4cfef

ggml : increased GGML_MAX_PARAMS to allow finetuning of 70b models (#4424)

Taikono-Himazin committed 1 year ago

Verified 41a11aaf

cmake : detect host compiler and cuda compiler separately

cebtenzzre committed 1 year ago

a81a34ad

cmake : silence linker check stdout

cebtenzzre committed 1 year ago

abacb278

make : honor NVCC, LLAMA_CUDA_CCBIN, NVCCFLAGS

cebtenzzre committed 1 year ago

88781479

make editorconfig checker happy

cebtenzzre committed 1 year ago

93ca80fa

make : detect host compiler and cuda compiler separately

cebtenzzre committed 1 year ago

91df2623

make : simplify nvcc flags

cebtenzzre committed 1 year ago

9b28f341

Update README.md (#4388)

y10ab1 committed 1 year ago

Verified 8a7b2fa5

grammar : revert the replacement of llama_token_to_piece with id_to_token (#4396)

AlienKevin committed 1 year ago

Verified e18f7345

sync : ggml (new ops, tests, backend, etc.) (#4359)

ggerganov committed 1 year ago

Verified fe680e3d

llama : per-layer KV cache + quantum K cache (#4309)

ggerganov committed 1 year ago

Verified bcc0eb45

train : fix #4227 (double free in examples/train-text-from-scratch/train-text-from-scratch.cpp) (#4351)

casavaca committed 1 year ago

Verified 81bc9214

server : recognize cache_prompt parameter in OAI API (#4347)

ggerganov committed 1 year ago

Verified 05cd6e50

common : fix compile warning

ggerganov committed 1 year ago

Verified caa92492

speculative : support `--color` (#4343)

stduhpf committed 1 year ago

Verified da5eaef1

grammar : pre-computed pieces + reserve mem + less string copies (#4330)

MarcusDunn committed 1 year ago

Verified 5f6e0c0d

Older