The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
pytorch xpu should be flash or mem efficient attention?
comfyanonymousb1fd26fe9e55
diff --git a/comfy/model_management.py b/comfy/model_management.pyindex 3b9fad362..a5142d305 100644--- a/comfy/model_management.py+++ b/comfy/model_management.py@@ -693,6 +693,8 @@ def pytorch_attention_flash_attention():#TODO: more reliable way of checking for flash attention?if is_nvidia(): #pytorch flash attention only works on Nvidiareturn True+ if is_intel_xpu():+ return Truereturn Falsedef force_upcast_attention_dtype():