fp8 casting is fast on GPUs that support fp8 compute.

comfyanonymous · Oct 20, 2024 · 471cd3e · 471cd3e
1 parent a68bbaf
commit 471cd3e
Showing 1 changed file with 3 additions and 0 deletions.
diff --git a/comfy/model_management.py b/comfy/model_management.py
@@ -647,6 +647,9 @@ def unet_dtype(device=None, model_params=0, supported_dtypes=[torch.float16, tor
  pass
 
  if fp8_dtype is not None:
+ if supports_fp8_compute(device): #if fp8 compute is supported the casting is most likely not expensive
+ return fp8_dtype
+
  free_model_memory = maximum_vram_for_weights(device)
  if model_params * 2 > free_model_memory:
  return fp8_dtype