Skip to content

Releases: ggml-org/llama.cpp

b5477

24 May 23:11
a2d02d5
Compare
Choose a tag to compare
releases : bundle llvm omp library in windows release (#13763)

b5476

24 May 20:54
17fc817
Compare
Choose a tag to compare
releases : enable openmp in windows cpu backend build (#13756)

b5475

24 May 20:44
2bd1b30
Compare
Choose a tag to compare
ggml-cpu : set openmp wait time if not set (#13758)

b5474

24 May 15:16
259469c
Compare
Choose a tag to compare
Move GLM4 f32 attention fix to the correct function (#13750)

b5473

24 May 11:32
4c32832
Compare
Choose a tag to compare
ggml : add ggml_gelu_erf() CUDA kernel (#13719)

* ggml : add ggml_gelu_erf() CUDA kernel

* missing semicolon

b5472

24 May 10:48
c3a2624
Compare
Choose a tag to compare
vocab : fix ugm tokenizer precision (#13743)

b5471

24 May 10:12
ffd0eae
Compare
Choose a tag to compare
CUDA: fix race condition in FA vector kernels (#13742)

b5468

23 May 17:34
d13d0f6
Compare
Choose a tag to compare
hparams : initialize arrays (#13728)

ggml-ci

b5466

23 May 10:29
9ecf3e6
Compare
Choose a tag to compare
server : support audio input (#13714)

* server : support audio input

* add audio support on webui

b5465

23 May 10:01
faaaff5
Compare
Choose a tag to compare
CANN: Support MUL_MAT_ID for q8_0 and q4_0 (#13705)

* [CANN]Support MUL_MAT_ID Q8 && Q4

Signed-off-by: noemotiovon <[email protected]>

* codestyle adjustment

Signed-off-by: noemotiovon <[email protected]>

---------

Signed-off-by: noemotiovon <[email protected]>