revisited this and love the refiner. Was doing the non-refiner a while for faster image generation, but with the refiner it adds so many nice details... Now coming back to this with dynamic prompts and a small army of text files to power it is creating amazing results. If you get in to dynamic prompts two tools I'm finding invaluable are https://textcleaner.net/ to remove extra characters and https://codverter.com/src/maineditor text splitter option to cut the text in to 1000 character chunks per line... that way python won't fumble over the special characters, etc. and the text is long enough on each line that it comes across as a full idea instead of just little scattered words.
It's worth noting, (btw i don't understand why my comments appear on the wrong page) - that with models like Juggernaut, a refiner is considered detrimental more than anything else. But given you were focused on SDXL itself, I get it. Great as usual!
That is the whole message. I can post a screenshot if you'd like. And yes, I loaded this with your png and didn't even change the prompt. This is literally my first run.
Sorry, for the trouble. Any advice on how to proceed?
revisited this and love the refiner. Was doing the non-refiner a while for faster image generation, but with the refiner it adds so many nice details... Now coming back to this with dynamic prompts and a small army of text files to power it is creating amazing results. If you get in to dynamic prompts two tools I'm finding invaluable are https://textcleaner.net/ to remove extra characters and https://codverter.com/src/maineditor text splitter option to cut the text in to 1000 character chunks per line... that way python won't fumble over the special characters, etc. and the text is long enough on each line that it comes across as a full idea instead of just little scattered words.
It's worth noting, (btw i don't understand why my comments appear on the wrong page) - that with models like Juggernaut, a refiner is considered detrimental more than anything else. But given you were focused on SDXL itself, I get it. Great as usual!
Any idea what to do with this?
Error occurred when executing CheckpointLoaderSimple:
Error while deserializing header: HeaderTooLarge
File "/workspace/ComfyUI/execution.py", line 151, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "/workspace/ComfyUI/execution.py", line 81, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "/workspace/ComfyUI/execution.py", line 74, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "/workspace/ComfyUI/nodes.py", line 446, in load_checkpoint
out = comfy.sd.load_checkpoint_guess_config(ckpt_path, output_vae=True, output_clip=True, embedding_directory=folder_paths.get_folder_paths("embeddings"))
File "/workspace/ComfyUI/comfy/sd.py", line 1307, in load_checkpoint_guess_config
sd = utils.load_torch_file(ckpt_path)
File "/workspace/ComfyUI/comfy/utils.py", line 11, in load_torch_file
sd = safetensors.torch.load_file(ckpt, device=device.type)
File "/workspace/ComfyUI/venv/lib/python3.10/site-packages/safetensors/torch.py", line 309, in load_file
with safe_open(filename, framework="pt", device=device) as f:
hmm, can you try if you can load the workflow with this image? just download and drag it onto your canvas https://huggingface.co/pxovela/sdxl_comfyui_post_series/resolve/main/27%20-%20ComfyUI%20explrotaion%20part%203/ComfyUI_temp_npftc_00121_.png
also, do you have the end of that error message? Typically you should read those erorrs from bottom up
That is the whole message. I can post a screenshot if you'd like. And yes, I loaded this with your png and didn't even change the prompt. This is literally my first run.
Sorry, for the trouble. Any advice on how to proceed?
weird. I suspect it's one of the two:
easier to fix: you don't have the right checkpoints in the folder?
Harder: something wrong with the installation as a whole and you need to fix that first if, in general, none of the workflows at all work for you.