absolutely, i'm running the full flux.1-dev model on an amd 7900xtx. i had to disable smart model loading in comfyui to avoid out-of-memory issues.
note: depending on whether you're using the official unet models from black forest labs, or a quantized checkpoint, the workflows are different. see: https://comfyui-wiki.com/tutorial/advanced/flux1-comfyui-gui...