Releases: wangqi/llama.cpp
Releases · wangqi/llama.cpp
b3756
Add stdbool.h to examples/llava/clip.h
b3755
ggml : ggml_type_name return "NONE" for invalid values (#9458) When running on Windows, the quantization utility attempts to print the types that are not set which leads to a crash.
b2886
script : sync ggml-rpc
b2830
CUDA: generalize FP16 fattn vec kernel (#7061) * CUDA: generalize FP16 fattn vec kernel * disable unsupported head sizes for AMD in test * try AMD fix * fix batch size 2-8 * partially revert changes