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.

I found a case where this is needed (#10875)

comfyanonymous015a0599d08f
diff --git a/comfy/quant_ops.py b/comfy/quant_ops.py
index e938144a7..0c16bcf8d 100644
--- a/comfy/quant_ops.py
+++ b/comfy/quant_ops.py
@@ -405,8 +405,8 @@ class TensorCoreFP8Layout(QuantizedLayout):
tensor_scaled = tensor * (1.0 / scale).to(tensor.dtype)
# TODO: uncomment this if it's actually needed because the clamp has a small performance penality'
- # lp_amax = torch.finfo(dtype).max
- # torch.clamp(tensor_scaled, min=-lp_amax, max=lp_amax, out=tensor_scaled)
+ lp_amax = torch.finfo(dtype).max
+ torch.clamp(tensor_scaled, min=-lp_amax, max=lp_amax, out=tensor_scaled)
qdata = tensor_scaled.to(dtype, memory_format=torch.contiguous_format)
layout_params = {