RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

[dnvs]

hope support mac ...

[Kiteretsu77]

Thanks for the report! I will try my best to add a feature to be capable of doing inference on CPU-only devices. Meanwhile, in ComfyUI-APISR, they have provided a way to use the CPU. However, this repo is in Chinese. You can have a look at that and it may be possible that it is adaptive to Mac.
Another solution is to use the colab provided to use free account credit from Google to do inference online. Thanks!

[alexaex]

I have resolved this issue. You just need to add a few lines of code like the screenshots. I've tried this and I'm able to successfully run the inference on a MacBook Air M2.