Skip to content

Releases: ServeurpersoCom/llama.cpp

b6627

29 Sep 10:37
adc7634
Compare
Choose a tag to compare
ggml : check cuda and metal argsort limits and add test (#16323)

* check cuda argsort limits and add test

* add metal check

b6625

29 Sep 07:33
66bb798
Compare
Choose a tag to compare
fix: preserved zero values in chat settings inputs and textareas by s…

b6622

29 Sep 06:04
a4a0aa5
Compare
Choose a tag to compare
ggml : fix dependencies for ggml_set_rows (#16318)

b6620

28 Sep 23:23
b887d2f
Compare
Choose a tag to compare
ggml : fix GGML_F32_VEC_FMA argument order in ggml_vec_mad1_f32 (#16307)

* fix GGML_F32_VEC_FMA argument order in ggml_vec_mad1_f32

* add test that fails on simd

b6619

28 Sep 18:34
bd0af02
Compare
Choose a tag to compare
common : fix reasoning before forced tool call via tool_choice = requ…

b6615

28 Sep 08:44
d8359f5
Compare
Choose a tag to compare
vulkan: 64-bit im2col (#16135)

* vulkan: 64-bit im2col

Add variants of the im2col shaders that use buffer_device_address/buffer_reference,
and use 64-bit address calculations. This is needed for large convolutions used in
stable-diffusion.cpp.

* fix validation error for large im2col