sync : llama.cpp #1315
ggerganov
merged
14 commits into
master
from
sync-llama.cpp-25-07-28
musa: upgrade musa sdk to rc4.2.0 (llama/14498)
b8cbf175
sched : fix multiple evaluations of the same graph with pipeline para…
d53fef6a
rpc : check for null buffers in get/set/copy tensor endpoints (llama/…
c7dd462f
ggml : remove invalid portPos specifiers from dot files (llama/14838)
4ad50948
opencl: add fused `rms_norm_mul` (llama/14841)
eab7bdb4
metal: SSM_SCAN performance (llama/14743)
afaae7cf
ggml-cpu : disable GGML_NNPA by default due to instability (llama/14880)
a6782ef9
musa: fix build warnings (unused variable) (llama/14869)
81f39973
CANN: Implement GLU ops (llama/14884)
ea155558
HIP: Enable Matrix cores for MMQ Kernels, Enable stream-K for CDNA 3 …
d743740d
Docs: add instructions for adding backends (llama/14889)
0d380cd7
vulkan: skip empty set_rows to avoid invalid API usage (llama/14860)
fccda47a
vulkan : add fp16 support for the conv_2d kernel (llama/14872)
62f0d682
sync : llama.cpp
8ca55dac
danbev
approved these changes
on 2025-07-28
Login to write a write a comment.
Login via GitHub