happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

GPU remove Q6_K requantization

This commit is contained in:
Yu, Zijun 2025-12-08 16:17:40 +08:00 committed by Mustafa Cavus
parent 28da9a9adc
commit 469325c6da
1 changed files with 0 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-openvino/utils.cpp
View File

@ -422,12 +422,6 @@ std::map<ggml_type, ExtraQuantType> get_types_to_requant(const std::string & dev
{GGML_TYPE_Q5_K, ExtraQuantType::F16 }, {GGML_TYPE_Q5_K, ExtraQuantType::F16 },
}; };
} }
if (device == "GPU") {
return {
// gs16 will be supported on openvino-2025.4
{GGML_TYPE_Q6_K, ExtraQuantType::Q8_0_32},
};
}
return {}; return {};
} }