Skip to content

LLM inference in C/C++

C++ ggml

Commits on master

hexagon: add q5_k quant type support (#29123)
Jhen-Jie Hong committed 1 month ago
4de0926
ed319fe
84e76d8
metal : optimize sparse FA + clean-up (#29377)
Georgi Gerganov committed 1 month ago
cdc0642
sync : ggml (#29396)
Georgi Gerganov committed 1 month ago
bced459
hexagon: handle multi-sequence in concat_2d (#29344)
Jhen-Jie Hong committed 1 month ago
a02c7f5
07fc586
hexagon: support I32 CPY and CONT (#29379)
kurquhar committed 1 month ago
97a418b
test: flush status (#28352)
kurquhar committed 1 month ago
8212c78
llama: add llama_batch_ext (#24669)
Xuan-Son Nguyen committed 1 month ago
fc343a8
server : change default pytest workers to 4 (#29376)
Daniel Bevenius committed 1 month ago
308883b
70596c4
6b790a9
53ed051
model : add Ling 3.0 VL support (#29151)
Toby committed 1 month ago
f830688
9710a32
bd4f514
server: allow preset to set log file (#29334)
Xuan-Son Nguyen committed 1 month ago
b9ae43a
fee39dd
llama.cpp : bump version to 0.5.0 (#29333)
Georgi Gerganov committed 1 month ago
7fe450e