Open guazike opened 8 months ago
Hello everyone, I have the same (first mentioned) problem as @guazike:
images are generated and saved in the corresponding folder (... \ComfyUI_windows_portable\ComfyUI\output\gligen), but the generated images are not displayed in the GLIGEN GUI.
I'm currently using the additional “comfyui-gligengui-node” to help me out and always switch back and forth between the two UIs, but it would be very helpful to see the images directly in GLIGEN-GUI :)
@mut-ex does it have anything to do with the new ComfyUI version?
My colleague found a solution:
you had to add the flag --enable-cors-header='http://localhost:8080'
when starting ComfyUI.
Starting ComfyUI Server with CORS Enabled to allow requests from ComfyUI (port 8080) for GLIGEN-GUI. In my case I had to edit the file “run_nvidia_gpu.bat” in my ComfyUI_windows_portable folder.
I have finish every steps in "Getting Started". When I draw a box just like in the screenshot and then click "QUEUE", gligen-gui notice "success". But the generated new image(found in ComfyUI/output/gligen/) not update in gligen-gui. And there's another problem: the gligen_sd14_textbox_pruned.safetensors dose not appear in the list of "CHECKPOINT", I'm sure I've put it in ComfyUI/models/gligen/ ![Uploading 截屏2024-03-08 18.16.17.png…]()