Skip to content

Releases: createthis/llama.cpp

b5939

19 Jul 16:19
f0d4d17
Compare
Choose a tag to compare
Documentation: Update build.md's Vulkan section (#14736)

* Documentation: Rewrote and updated the "Without docker" portion of the Vulkan backend build documentation.

* Documentation: Reorganize build.md's Vulkan section.

b5897

15 Jul 01:23
bdca383
Compare
Choose a tag to compare
sycl: Hotfix for non dnnl codepath (#14677)

b5833

05 Jul 16:31
a0374a6
Compare
Choose a tag to compare
vulkan: Handle updated FA dim2/3 definition (#14518)

* vulkan: Handle updated FA dim2/3 definition

Pack mask boolean and n_head_log2 into a single dword to keep the push
constant block under the 128B limit.

* handle null mask for gqa

* allow gqa with dim3>1

b5672

15 Jun 19:56
30e5b01
Compare
Choose a tag to compare
quantize : change int to unsigned int for KV overrides (#14197)

b5606

09 Jun 03:35
247e5c6
Compare
Choose a tag to compare
cuda : fix buffer type check with integrated GPUs (#14069)

b5602

07 Jun 03:00
745aa53
Compare
Choose a tag to compare
llama : deprecate llama_kv_self_ API (#14030)

* llama : deprecate llama_kv_self_ API

ggml-ci

* llama : allow llama_memory_(nullptr)

ggml-ci

* memory : add flag for optional data clear in llama_memory_clear

ggml-ci