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.

ops: tolerate already force casted dynamic weight (#14410)

rattusbda19b260488
diff --git a/comfy/ops.py b/comfy/ops.py
index 119177c37..3c9912aae 100644
--- a/comfy/ops.py
+++ b/comfy/ops.py
@@ -299,21 +299,21 @@ def cast_bias_weight(s, input=None, dtype=None, device=None, bias_dtype=None, of
non_blocking = comfy.model_management.device_supports_non_blocking(device)
- if hasattr(s, "_v"):
+ if hasattr(s, "_v") and comfy.model_management.is_device_cpu(device):
#vbar doesn't support CPU weights, but some custom nodes have weird paths
#that might switch the layer to the CPU and expect it to work. We have to take
#a clone conservatively as we are mmapped and some SFT files are packed misaligned
#If you are a custom node author reading this, please move your layer to the GPU
#or declare your ModelPatcher as CPU in the first place.
- if comfy.model_management.is_device_cpu(device):
- materialize_meta_param(s, ["weight", "bias"])
- weight = s.weight.to(dtype=dtype, copy=True)
- if isinstance(weight, QuantizedTensor):
- weight = weight.dequantize()
- bias = s.bias.to(dtype=bias_dtype, copy=True) if s.bias is not None else None
- return format_return((weight, bias, (None, None, None)), offloadable)
+ materialize_meta_param(s, ["weight", "bias"])
+ weight = s.weight.to(dtype=dtype, copy=True)
+ if isinstance(weight, QuantizedTensor):
+ weight = weight.dequantize()
+ bias = s.bias.to(dtype=bias_dtype, copy=True) if s.bias is not None else None
+ return format_return((weight, bias, (None, None, None)), offloadable)
+ elif hasattr(s, "_v") and s.weight.device != device:
prefetched = hasattr(s, "_prefetch")
offload_stream = None
offload_device = None