Releases: ngxson/llama.cpp
Releases · ngxson/llama.cpp
b5966
b5965
CUDA: add fused rms norm (#14800)
b5963
vulkan: fix rms_norm_mul to handle broadcasting dim0 (#14817)
b5962
llama : add model type detection for rwkv7 7B&14B (#14816) Signed-off-by: Molly Sophia <[email protected]>
b5961
imatrix: add option to display importance score statistics for a give…
b5960
Mtmd: add a way to select device for vision encoder (#14236) * Mtmd: add a way to select device for vision encoder * simplify * format * Warn user if manual device selection failed * initialize backend to nullptr
b5959
cuda : implement bf16 cpy ops and enable bf16 cont (#14763) * implement bf16 cpy ops and enable bf16 cont * deduplicate copy functions * deduplicate checks
b5958
opencl: remove unreachable `return` (#14806)
b5957
server : allow setting `--reverse-prompt` arg (#14799) Signed-off-by: Molly Sophia <[email protected]>
b5956
cuda: remove linking to cublasLt (#14790) Signed-off-by: Xiaodong Ye <[email protected]>