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.

Make the casting in lists the same as regular inputs. (#8373)

comfyanonymousfb4754624d0f
diff --git a/comfy/model_base.py b/comfy/model_base.py
index 638b04092..e0c2bcaa8 100644
--- a/comfy/model_base.py
+++ b/comfy/model_base.py
@@ -102,6 +102,13 @@ def model_sampling(model_config, model_type):
return ModelSampling(model_config)
+def convert_tensor(extra, dtype):
+ if hasattr(extra, "dtype"):
+ if extra.dtype != torch.int and extra.dtype != torch.long:
+ extra = extra.to(dtype)
+ return extra
+
+
class BaseModel(torch.nn.Module):
def __init__(self, model_config, model_type=ModelType.EPS, device=None, unet_model=UNetModel):
super().__init__()
@@ -165,13 +172,13 @@ class BaseModel(torch.nn.Module):
extra_conds = {}
for o in kwargs:
extra = kwargs[o]
+
if hasattr(extra, "dtype"):
- if extra.dtype != torch.int and extra.dtype != torch.long:
- extra = extra.to(dtype)
- if isinstance(extra, list):
+ extra = convert_tensor(extra, dtype)
+ elif isinstance(extra, list):
ex = []
for ext in extra:
- ex.append(ext.to(dtype))
+ ex.append(convert_tensor(ext, dtype))
extra = ex
extra_conds[o] = extra