mirrors/ollama-for-amd
1
0
Fork 0
mirror of https://github.com/likelovewant/ollama-for-amd.git synced 2025-12-21 22:33:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

vulkan: enable flash attention (#12937)

Browse Source 
Also adjusts the vulkan windows build pattern to match recent changes in other backends
so incremental builds are faster.
This commit is contained in:
Daniel Hiltgen
2025-11-04 10:31:22 -08:00
committed by GitHub
parent ef549d513c
commit a4770107a6
2 changed files with 5 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
ml/device.go
View File

@@ -432,7 +432,8 @@ func FlashAttentionSupported(l []DeviceInfo) bool {
supportsFA := gpu.Library == "cpu" ||
gpu.Name == "Metal" || gpu.Library == "Metal" ||
(gpu.Library == "CUDA" && gpu.DriverMajor >= 7 && !(gpu.ComputeMajor == 7 && gpu.ComputeMinor == 2)) ||
gpu.Library == "ROCm"
gpu.Library == "ROCm" ||
gpu.Library == "Vulkan"
if !supportsFA {
return false