Skip to content

Releases: wangqi/llama.cpp

b3756

14 Sep 19:38
Compare
Choose a tag to compare
Add stdbool.h to examples/llava/clip.h

b3755

14 Sep 19:28
822b632
Compare
Choose a tag to compare
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

14 May 19:30
9f77348
Compare
Choose a tag to compare
script : sync ggml-rpc

b2830

09 May 13:59
a743d76
Compare
Choose a tag to compare
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