COMMITS
August 24, 2026
J
convert: fix GLM regression in index_tensors (#27655)
jacekpoplawski committed
G
ggml : shorten virtual device naming in CUDA and Metal (#27608)
Georgi Gerganov committed
F
X
mtmd: video: fix moov atom at the end of file (#27596)
Xuan-Son Nguyen committed
G
ci : apply ccache-clear with older/min/dry-run to all ccache jobs (#27602)
Georgi Gerganov committed
G
ggml : fix ggml_clamp (#27644)
Georgi Gerganov committed
P
mamba2 : Flatten in/out projections to dispatch GEMM instead of GEMV (#27513)
Prabhsimran Singh committed
A
Deepseek 4: `-sm tensor` (#26490)
Aman Gupta committed
August 23, 2026
J
model : support MTP in GLM-4.5-Air (#26534)
jacekpoplawski committed
G
readme : update links (#27617)
Georgi Gerganov committed
A
fix: Change chat tabs nav shortcuts (#27609)
Aleksander Grygier committed
G
test : fix multi-GPU server tests (#27614)
Georgi Gerganov committed
X
test: move tools/parser to tests (#27548)
Xuan-Son Nguyen committed
X
mtmd: use pillow-accurate algo, correct resize_algo for all models (#27594)
Xuan-Son Nguyen committed
G
ci : add test-llama-archs tensor split for Metal (#27598)
Georgi Gerganov committed
N
contrib : recommend waiting for CI before merging (#27603)
Niklas Wenzel committed
G
server : add LLAMA_SERVER_SLOTS_N_DIFF (#27600)
Georgi Gerganov committed
B
common : skip device_info loop if it's not going to be printed (#26692)
Bartosz Taudul committed
A
DeepseekV4: fix rollback with multi-seq (#26756)
Aman Gupta committed
G
[Tensor parallel] Fix meta tensor split state propagation (#27574)
Gaurav Garg committed
A
ui: Chat Conversation Tabbed navigation (#27263)
Aleksander Grygier committed
A
vendor : update subprocess.h (#27409)
Alessandro de Oliveira Faria (A.K.A.CABELO) committed
A
cuda : add POOL_1D support (#27573)
Aman Karki committed
August 22, 2026
X
common: json.h: fix clang lto (#27575)
Xuan-Son Nguyen committed
S
vulkan : added the PAD_REFLECT_1D operation (#26586)
Safi Ullah committed
X
mtmd: use ggml_rope_set_offset (#27521)
Xuan-Son Nguyen committed
X
common: add json.h abstraction (#27511)
Xuan-Son Nguyen committed
X
fit: also take into account n_streams (#27496)
Xuan-Son Nguyen committed
X
arg: remove -no-cnv from cli [no ci] (#27542)
Xuan-Son Nguyen committed
M
ci : Restore ROCm job for Ubuntu (#27399)
Mario Limonciello committed