root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-02 14:54:35 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

kompute : remove Q6_K from list of supported quant types

Browse Source
This commit is contained in:
Cebtenzzre 2023-10-04 16:19:19 -04:00 committed by cebtenzzre
parent 24a4a5956a
commit 3d850db767
1 changed files with 1 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
llama.cpp
View File

@ -6498,8 +6498,7 @@ struct llama_context * llama_new_context_with_model(
&& (model->ftype == LLAMA_FTYPE_ALL_F32 && (model->ftype == LLAMA_FTYPE_ALL_F32
|| model->ftype == LLAMA_FTYPE_MOSTLY_F16 || model->ftype == LLAMA_FTYPE_MOSTLY_F16
|| model->ftype == LLAMA_FTYPE_MOSTLY_Q4_0 || model->ftype == LLAMA_FTYPE_MOSTLY_Q4_0
|| model->ftype == LLAMA_FTYPE_MOSTLY_Q4_1 || model->ftype == LLAMA_FTYPE_MOSTLY_Q4_1)) {
|| model->ftype == LLAMA_FTYPE_MOSTLY_Q6_K)) {
// this allocates all Vulkan resources and memory buffers // this allocates all Vulkan resources and memory buffers
ctx->ctx_kompute = ggml_vk_init(); ctx->ctx_kompute = ggml_vk_init();