Skip to content

LLM inference in C/C++

C++ ggml

Commits on master

d4365d9
ci : add self-hosted webgpu to hf-jobs (#28712)
Sigbjørn Skjæret committed 2 months ago
583926e
hex-cpy: use dma if src and dst are contiguous (#28906)
Trivikram Reddy committed 2 months ago
72b590d
HIP: Enable AllReduce for ROCm (#27825)
Sandro Steeger committed 2 months ago
38a5b42
ci: build MUSA for only 1 arch (#28944)
Aman Gupta committed 2 months ago
d1d3c33
docs: Rule of thumb for AI review time [no ci] (#28945)
Johannes Gäßler committed 2 months ago
6011c34
rpc : hash-cache only weights (#28789)
Patrick Hoffmann committed 2 months ago
7609846
cuda: support row-contiguous SUM_ROWS (#26308)
Mohamed Elashri committed 2 months ago
5431581
vulkan: support sparse Flash Attention (#28105)
Ruben Ortlam committed 2 months ago
fc82583
opencl: add generic ssm_scan (#28881)
lhez committed 2 months ago
6ec1a7e
1af6c65
1e7bcf3
ci: Bump CUDA Windows x64 builds to 13.4.1 (#28930)
shivamkumard-ctrl committed 2 months ago
0ecb159
ci : fix android release (#28936)
Sigbjørn Skjæret committed 2 months ago
987498f
cuda : enable i16 and i32 for DUP (#28897)
Aman Karki committed 2 months ago
4c9233c
69eb250
7cf1c54
ci : add ubuntu-cuda builds to release (#28186)
Oliver Simons committed 2 months ago
391fac1
qwen4exp: enable rms_norm + mul fusion (#28896)
Aman Gupta committed 2 months ago
41abbfd
release : added gfx1103 to ubuntu rocm build (#28423)
Apoorv Parle committed 2 months ago
b4fa47d
cmake : remove precompiled headers (#28892)
Daniel Bevenius committed 2 months ago
f3a184b