adding Lora freezes "Model Initializing ... "?
The w4a8 model works fine but after loading Lora (with basic Load Lora node) it freezes loading the model and lora both. do i need to use a special custom node for lora? or is there a possible fix for this issue?
this is the current issue with w4a8, that will be fixed later
It should already be fixed in latest version, I'm not aware of further issues or experiencing any myself.
It should already be fixed in latest version, I'm not aware of further issues or experiencing any myself.
Which version should I update to?
Yeah also stuck on "Model Initializing" i dont know whats the problem also since there no crash i cant gather more info im using colab t4 nvidia turing
This still needs pytorch with cu130 or it will work very poorly.
The latest stable comfyui and the included comfy-kitchen (v0.2.28) includes the preliminary LoRA application optimizations, it is working for me and some others that have been testing, so make sure all that is in order.
This still needs pytorch with cu130 or it will work very poorly.
The latest stable comfyui and the included comfy-kitchen (v0.2.28) includes the preliminary LoRA application optimizations, it is working for me and some others that have been testing, so make sure all that is in order.
I try it and work well,thanks!!!
thanks! updating to 0.31.0 fixed it.
i tried your distilled lora it works fantastic better then all the rest