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

Delete TODO now that we have q8_0.

Browse Source
This commit is contained in:
Adam Treat 2023-10-05 10:52:04 -04:00 committed by cebtenzzre
parent 8564f79036
commit 09d83f0401
1 changed files with 0 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
llama.cpp
View File

@ -6510,7 +6510,6 @@ struct llama_context * llama_new_context_with_model(
#undef LLAMA_METAL_CHECK_BUF #undef LLAMA_METAL_CHECK_BUF
} }
#elif defined(GGML_USE_KOMPUTE) #elif defined(GGML_USE_KOMPUTE)
// TODO(cebtenzzre): we need to check the type of each tensor because Q8_0 is not currently supported
if (ggml_vk_has_device() && params.n_gpu_layers > 0 if (ggml_vk_has_device() && params.n_gpu_layers > 0
&& (model->arch == LLM_ARCH_LLAMA || model->arch == LLM_ARCH_FALCON) && (model->arch == LLM_ARCH_LLAMA || model->arch == LLM_ARCH_FALCON)
&& (model->ftype == LLAMA_FTYPE_ALL_F32 && (model->ftype == LLAMA_FTYPE_ALL_F32