The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.
Remove scaled_fp8 key after reading it to silence warning.
comfyanonymous5e29e7a488b3
diff --git a/comfy/model_detection.py b/comfy/model_detection.pyindex 229fe499d..b98820d83 100644--- a/comfy/model_detection.py+++ b/comfy/model_detection.py@@ -321,8 +321,9 @@ def model_config_from_unet(state_dict, unet_key_prefix, use_base_if_no_match=Falif model_config is None and use_base_if_no_match:model_config = comfy.supported_models_base.BASE(unet_config)- scaled_fp8_weight = state_dict.get("{}scaled_fp8".format(unet_key_prefix), None)- if scaled_fp8_weight is not None:+ scaled_fp8_key = "{}scaled_fp8".format(unet_key_prefix)+ if scaled_fp8_key in state_dict:+ scaled_fp8_weight = state_dict.pop(scaled_fp8_key)model_config.scaled_fp8 = scaled_fp8_weight.dtypeif model_config.scaled_fp8 == torch.float32:model_config.scaled_fp8 = torch.float8_e4m3fn