The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
Pass extra conds directly to unet.
comfyanonymousd1d2fea806c0
diff --git a/comfy/model_base.py b/comfy/model_base.pyindex edc246f8c..ea3ea61f2 100644--- a/comfy/model_base.py+++ b/comfy/model_base.py@@ -50,7 +50,7 @@ class BaseModel(torch.nn.Module):self.register_buffer('alphas_cumprod', torch.tensor(alphas_cumprod, dtype=torch.float32))self.register_buffer('alphas_cumprod_prev', torch.tensor(alphas_cumprod_prev, dtype=torch.float32))- def apply_model(self, x, t, c_concat=None, c_crossattn=None, c_adm=None, control=None, transformer_options={}, **kwargs):+ def apply_model(self, x, t, c_concat=None, c_crossattn=None, control=None, transformer_options={}, **kwargs):if c_concat is not None:xc = torch.cat([x] + [c_concat], dim=1)else:@@ -60,9 +60,10 @@ class BaseModel(torch.nn.Module):xc = xc.to(dtype)t = t.to(dtype)context = context.to(dtype)- if c_adm is not None:- c_adm = c_adm.to(dtype)- return self.diffusion_model(xc, t, context=context, y=c_adm, control=control, transformer_options=transformer_options).float()+ extra_conds = {}+ for o in kwargs:+ extra_conds[o] = kwargs[o].to(dtype)+ return self.diffusion_model(xc, t, context=context, control=control, transformer_options=transformer_options, **extra_conds).float()def get_dtype(self):return self.diffusion_model.dtype@@ -107,7 +108,7 @@ class BaseModel(torch.nn.Module):out['c_concat'] = comfy.conds.CONDNoiseShape(data)adm = self.encode_adm(**kwargs)if adm is not None:- out['c_adm'] = comfy.conds.CONDRegular(adm)+ out['y'] = comfy.conds.CONDRegular(adm)return outdef load_model_weights(self, sd, unet_prefix=""):