The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
fix: use Parameter assignment for Stable_Zero123 cc_projection weights (fixes #13492) (#13518)
Octopuscc6f9500a1b9
diff --git a/comfy/model_base.py b/comfy/model_base.pyindex 5c2668ba9..1c7695761 100644--- a/comfy/model_base.py+++ b/comfy/model_base.py@@ -578,8 +578,8 @@ class Stable_Zero123(BaseModel):def __init__(self, model_config, model_type=ModelType.EPS, device=None, cc_projection_weight=None, cc_projection_bias=None):super().__init__(model_config, model_type, device=device)self.cc_projection = comfy.ops.manual_cast.Linear(cc_projection_weight.shape[1], cc_projection_weight.shape[0], dtype=self.get_dtype(), device=device)- self.cc_projection.weight.copy_(cc_projection_weight)- self.cc_projection.bias.copy_(cc_projection_bias)+ self.cc_projection.weight = torch.nn.Parameter(cc_projection_weight.clone())+ self.cc_projection.bias = torch.nn.Parameter(cc_projection_bias.clone())def extra_conds(self, **kwargs):out = {}