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.

mix_ascend_bf16_infer_err (#6794)

zhoufan295635740259de27
diff --git a/comfy/model_management.py b/comfy/model_management.py
index 29cd43b51..cbf4c4ea6 100644
--- a/comfy/model_management.py
+++ b/comfy/model_management.py
@@ -1082,6 +1082,9 @@ def should_use_bf16(device=None, model_params=0, prioritize_performance=True, ma
if is_intel_xpu():
return True
+
+ if is_ascend_npu():
+ return True
props = torch.cuda.get_device_properties(device)
if props.major >= 8: