 Georgi Gerganov
|
cce3dcffc5
cuda : non-cont concat support (#7610)
|
1 year ago |
 Radoslav Gerganov
|
210d99173d
llama-bench : add support for the RPC backend (#7435)
|
1 year ago |
 slaren
|
87bdf2a199
ggml : use atomic_flag for critical section (#7598)
|
1 year ago |
|
Georgi Gerganov
|
00281b7be3
scripts : remove mpi remnants
|
1 year ago |
|
Georgi Gerganov
|
2ab977282b
sync : ggml
|
1 year ago |
|
Georgi Gerganov
|
72de268bec
ggml : restore ggml_rope_xpos_inplace (ggml/0)
|
1 year ago |
 Akarshan Biswas
|
0e8d8bfd6c
Add Arc A750 and Arch linux to readme-sycl.md as verified GPU model and Linux distro (#7605)
|
1 year ago |
 zhouwg
|
504f0c340f
ggml : fix typo in ggml.c (#7603)
|
1 year ago |
 Meng, Hengyu
|
b864b50ce5
[SYCL] Align GEMM dispatch (#7566)
|
1 year ago |
 jaime-m-p
|
02c1ecad07
Tokenizer WPM fixes (#7500)
|
1 year ago |
|
Georgi Gerganov
|
6bd12ce409
sycl : fix assert (#7563)
|
1 year ago |
 Giuseppe Scrivano
|
5442939fcc
llama : support small Granite models (#7481)
|
1 year ago |
 k.h.lai
|
56411a950f
vulkan: properly initialize vulkan devices for LLAMA_SPLIT_MODE_NONE (#7552)
|
1 year ago |
|
Radoslav Gerganov
|
2b737caae1
rpc : resource management rework (#7562)
|
1 year ago |
 fairydreaming
|
ee3dff6b8e
Add support for DeepseekV2ForCausalLM (#7519)
|
1 year ago |
|
Georgi Gerganov
|
edc29433fa
tests : fix test-tokenizer-0.sh
|
1 year ago |
|
Georgi Gerganov
|
8b99e2aa66
llama : handle unknown utf8 bytes (#7588)
|
1 year ago |
 Brian
|
271ff3fc44
github: add refactor to issue template (#7561)
|
1 year ago |
 Neo Zhang
|
e2b065071c
[SYCL]fix ggml_sycl_mul_mat_id() to match the change of api (#7436)
|
1 year ago |
|
Georgi Gerganov
|
0548a4187f
ggml : generalize GGML_OP_CONCAT (#7563)
|
1 year ago |
 mgroeber9110
|
9335b969e8
server: do not remove whitespace at the start of a completion chunk (#7524)
|
1 year ago |
 Nathan Epstein
|
c41767154e
Markdownish code block fix (#7571)
|
1 year ago |
 Ikko Eltociear Ashimine
|
74b239b3d5
llava : update clip.h (#7580)
|
1 year ago |
 Djip007
|
852aafb163
update HIP_UMA #7399 (#7414)
|
1 year ago |
 kunnis
|
0136966daf
adding in x64 targets to cmake presets (#7574)
|
1 year ago |
 Johannes Gäßler
|
10b1e45876
make: add --device-debug to NVCC debug flags (#7542)
|
1 year ago |
 agray3
|
197c00681b
Allow multiple copy function pointers for CUDA graph kernel param updates (#7565)
|
1 year ago |
 AidanBeltonS
|
95f84d5ce8
Fix q_xxs using mul_mat_q (#7459)
|
1 year ago |
|
AidanBeltonS
|
5487593bc7
Add freq factors (#7495)
|
1 year ago |
|
Georgi Gerganov
|
1d8fca72ae
metal : add GGML_OP_REPEAT kernels (#7557)
|
1 year ago |
