vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

iq1_s: CUDA is working

Browse source
This commit is contained in:
Iwan Kawrakow 2024-02-11 13:08:26 +02:00
parent 80cd5bae99
commit a9d48e9718
6 changed files with 412 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

6
ggml.c
View file

@ -19071,8 +19071,10 @@ void ggml_quantize_init(enum ggml_type type) {
void ggml_quantize_free(void) {
ggml_critical_section_start();
iq2xs_free_impl(256);
iq2xs_free_impl(512);
iq2xs_free_impl(GGML_TYPE_IQ2_XXS);
iq2xs_free_impl(GGML_TYPE_IQ2_XS);
iq2xs_free_impl(GGML_TYPE_IQ1_S);
iq3xs_free_impl(256);
ggml_critical_section_end();
}