Skip to content

The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.

Enable pytorch attention in the VAE on AMD (#17271)

DELUXAbef11fcbfa8d
diff --git a/comfy/model_management.py b/comfy/model_management.py
index aa3e04822..3d639e099 100644
--- a/comfy/model_management.py
+++ b/comfy/model_management.py
@@ -1776,8 +1776,6 @@ def pytorch_attention_enabled():
return ENABLE_PYTORCH_ATTENTION
def pytorch_attention_enabled_vae():
- if is_amd():
- return False # enabling pytorch attention on AMD currently causes crash when doing high res
return pytorch_attention_enabled()
def pytorch_attention_flash_attention():