The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
Support MT5.
comfyanonymous0a4c49c57ca7
diff --git a/comfy/text_encoders/t5.py b/comfy/text_encoders/t5.pyindex 448c5aad3..ba628478d 100644--- a/comfy/text_encoders/t5.py+++ b/comfy/text_encoders/t5.py@@ -223,7 +223,7 @@ class T5(torch.nn.Module):self.num_layers = config_dict["num_layers"]model_dim = config_dict["d_model"]- self.encoder = T5Stack(self.num_layers, model_dim, model_dim, config_dict["d_ff"], config_dict["dense_act_fn"], config_dict["is_gated_act"], config_dict["num_heads"], config_dict["model_type"] == "t5", dtype, device, operations)+ self.encoder = T5Stack(self.num_layers, model_dim, model_dim, config_dict["d_ff"], config_dict["dense_act_fn"], config_dict["is_gated_act"], config_dict["num_heads"], config_dict["model_type"] != "umt5", dtype, device, operations)self.dtype = dtypeself.shared = torch.nn.Embedding(config_dict["vocab_size"], model_dim, device=device)