8000 Vulkan: Add f32 accumulator support to quantized mul mat to fix GLM4 … · robbiemu/llama.cpp@8960efd · GitHub
[go: up one dir, main page]
Skip to content

Commit 8960efd

Browse files
0cc4m0cc4m
authored
Vulkan: Add f32 accumulator support to quantized mul mat to fix GLM4 32B incoherence (ggml-org#13607)
1 parent 725f23f commit 8960efd

File tree

1 file changed

+118
-108
lines changed

1 file changed

+118
-108
lines changed

0 commit comments

Comments
 (0)
0