 Johannes Gäßler
|
963552903f
CUDA: fix broken oob check for FA vec f32 kernel (#7904)
|
1 anno fa |
 Georgi Gerganov
|
a9cae48003
tests : add non-cont unary tests (#7857)
|
1 anno fa |
|
Georgi Gerganov
|
bfaa676b08
ggml : improve ggml_is_contiguous logic (#7856)
|
1 anno fa |
|
Georgi Gerganov
|
704a35b183
server : restore numeric prompts (#7883)
|
1 anno fa |
 Meng, Hengyu
|
dcf752707d
update intel docker oneapi-basekit to 2024.1.1-devel-ubuntu22.04 (#7894)
|
1 anno fa |
 Patrice Ferlet
|
f2b5764beb
Fix a typo and add Fedora 40 pacakge to install for Vulkan (#7794) [no ci]
|
1 anno fa |
 k.h.lai
|
73bac2b11d
vulkan: select only one device for single gpu with multiple drivers (#7582)
|
1 anno fa |
 0cc4m
|
ef52d1d16a
Update Vulkan RoPE implementation (#7818)
|
1 anno fa |
 Deven Mistry
|
14f83526cd
fix broken link in pr template (#7880) [no ci]
|
1 anno fa |
 Brian
|
6fe42d073f
github: move PR template to .github/ root (#7868)
|
1 anno fa |
|
Johannes Gäßler
|
148995e5e5
llama-bench: more compact markdown tables (#7879)
|
1 anno fa |
|
Georgi Gerganov
|
4bfe50f741
tests : check the Python version (#7872)
|
1 anno fa |
|
Johannes Gäßler
|
bdcb8f4222
CUDA: int8 tensor cores for MMQ (q4_K, q5_K, q6_K) (#7860)
|
1 anno fa |
 slaren
|
c2ce6c47e4
fix CUDA CI by using a windows-2019 image (#7861)
|
1 anno fa |
 Olivier Chafik
|
b61eb9644d
json: refine constraint for whitespace to avoid runaways yet allow pretty print (#7866)
|
1 anno fa |
|
Olivier Chafik
|
396b18dfec
`json`: document schema conversion in GBNF readme, align manual grammar examples & converters (#7841)
|
1 anno fa |
 Jared Van Bortel
|
864a99e7a0
cmake : fix CMake requirement for CUDA (#7821)
|
1 anno fa |
|
slaren
|
fd5ea0f897
ci : try win-2019 on server windows test (#7854)
|
1 anno fa |
|
Georgi Gerganov
|
c28a83902c
examples : remove --instruct remnants (#7846)
|
1 anno fa |
|
Georgi Gerganov
|
d9da0e4986
server : improve "prompt" handling (#7847)
|
1 anno fa |
|
Johannes Gäßler
|
1f0dabda8d
CUDA: use tensor cores for MMQ (#7676)
|
1 anno fa |
 Ben Ashbaugh
|
af4ae502dd
use the correct SYCL context for host USM allocations (#7777)
|
1 anno fa |
|
Georgi Gerganov
|
10ceba354a
flake.lock: Update (#7838)
|
1 anno fa |
|
Georgi Gerganov
|
e95beeb1fc
imatrix : handle partial entries (#7833)
|
1 anno fa |
 Nicolás Pérez
|
57bf62ce7c
docs: Added initial PR template with directions for doc only changes and squash merges [no ci] (#7700)
|
1 anno fa |
 mgroeber9110
|
3e2ee44315
server: do not remove whitespace at the start of a completion chunk (#7830)
|
1 anno fa |
|
Johannes Gäßler
|
42b53d192f
CUDA: revise q8_1 data layout for mul_mat_q (#7824)
|
1 anno fa |
 sasha0552
|
2decf57bc6
convert-hf : set the model name based on cli arg, if present (#7693)
|
1 anno fa |
 compilade
|
5795b94182
convert-hf : match model part name prefix and suffix (#7687)
|
1 anno fa |
|
compilade
|
ed9f252118
gguf-py : decouple adding metadata from writing in GGUFWriter (#7827)
|
1 anno fa |
