Is there a way to use other lightweight models here

[libai-lab]

Is there a way to use other lightweight models here,such as Lykon/dreamshaper-xl-v2-turbo,Or other models can be faster and lighter, can openvino be added

[rupeshs]

Other models also possible I will release LCM version but it will be slower 3 steps minimum. OpenVINO is taking too much memory around 30GB RAM required I'm thinking about other optimizations. Currently we are using sdxl turbo.

[libai-lab]

Other models also possible I will release LCM version but it will be slower 3 steps minimum. OpenVINO is taking too much memory around 30GB RAM required I'm thinking about other optimizations. Currently we are using sdxl turbo.

How long does it take to generate the current version? Why use execution to controllnet=ControlNetModel.free_pretrained in codespace（

Controllnet_path,

Torch_dtype=torch_dtype,

）Terminated, is the memory too small? I used it under the CPU of Colab, and the generation was very slow and unsuccessful