 Johannes Gäßler
|
a20f3c7465
CUDA: fix tensor core logic for Pascal and HIP (#4682)
|
пре 2 година |
 Georgi Gerganov
|
0235b9b571
clip : use ggml_backend_buffer_is_host (#4205)
|
пре 2 година |
 Steward Garcia
|
ce18d727a4
clip : enable gpu backend (#4205)
|
пре 2 година |
 hydai
|
91bb39cec7
cuda: fix vmm oom issue on NVIDIA AGX Orin (#4687)
|
пре 2 година |
 crasm
|
04ac0607e9
python : add check-requirements.sh and GitHub workflow (#4585)
|
пре 2 година |
 Philip Taron
|
68eccbdc5b
flake.nix : rewrite (#4605)
|
пре 2 година |
 Cuong Trinh Manh
|
97bbca6e85
cmake : fix ld warning duplicate libraries libllama.a (#4671)
|
пре 2 година |
 Justine Tunney
|
4af4801566
llava-cli : refactor to use sampling library (#4669)
|
пре 2 година |
|
Justine Tunney
|
db49ff8ed7
server : replace sleep with condition variables (#4673)
|
пре 2 година |
 SakuraUmi
|
60f55e888c
server : fix OpenAI server sampling w.r.t. penalty. (#4675)
|
пре 2 година |
 Karthik Sethuraman
|
b93edd22f5
server : allow to generate multimodal embeddings (#4681)
|
пре 2 година |
 andrijdavid
|
82d6eab224
main-cmake-pkg : fix build issue (#4665)
|
пре 2 година |
 Peter Sugihara
|
afd997ab60
llama.swiftui : fix infinite loop, ouput timings, buff UI (#4674)
|
пре 2 година |
|
Georgi Gerganov
|
c8255f8a6b
scripts : print list of sync commits
|
пре 2 година |
 Tamotsu Takahashi
|
441f51dca0
ci : build with CLBlast + ggml-opencl use GGML_API (whisper/1576)
|
пре 2 година |
|
Georgi Gerganov
|
38b3de4658
sync : ggml
|
пре 2 година |
 bssrdf
|
afc8c19291
ggml : fix some mul mat cases + add tests for src1 F16 (ggml/669)
|
пре 2 година |
|
Georgi Gerganov
|
ca38b8d334
scripts : do not sync commits from this repo
|
пре 2 година |
|
Justine Tunney
|
65e5f6dadb
Fix OpenAI server sampling w.r.t. temp and seed (#4668)
|
пре 2 година |
 manikbhandari
|
ea5497df5d
gpt2 : Add gpt2 architecture integration (#4555)
|
пре 2 година |
 Nam D. Tran
|
f6793491b5
llama : add AWQ for llama, llama2, mpt, and mistral models (#4593)
|
пре 2 година |
 Daniel Bevenius
|
879b690a9e
finetune : fix output formatting in print_params (#4653)
|
пре 2 година |
|
Georgi Gerganov
|
b47879b0dd
scripts : add sync-ggml-am.sh
|
пре 2 година |
|
Georgi Gerganov
|
951010fa53
ggml : fix dot product for ARM (#4630)
|
пре 2 година |
 wonjun Jang
|
f56d6077d0
Add byte token type when tokenizer.model is not exists (#4641)
|
пре 2 година |
 slaren
|
dc68f0054c
cuda : fix vmm pool with multi GPU (#4620)
|
пре 2 година |
 WillCorticesAI
|
de8e496437
Update comment for AdamW implementation reference. (#4604)
|
пре 2 година |
 FantasyGmm
|
77465dad48
Fix new CUDA10 compilation errors (#4635)
|
пре 2 година |
 Paul Tsochantaris
|
a206137f92
Adding Emeltal reference to UI list (#4629)
|
пре 2 година |
|
slaren
|
b9f47952ff
simplify bug issue template (#4623)
|
пре 2 година |
