A collection of ComfyUI custom nodes to help streamline workflows and reduce total node count.
Github Repo: https://github.com/LucianoCirino/efficiency-nodes-comfyui
Currently Available Nodes:
Ksampler (Efficient)
A modded KSampler with the ability to preview/output images and run scripts.
Efficient Loader
A combination of common initialization nodes.
XY Plotter Nodes
A collection of nodes that allows users to specify parameters for the KSampler (Efficient) to plot on a grid.
Image Overlay
Node that allows for flexible image overlaying.
SimpleEval Nodes
A collection of nodes that allows users to write simple Python expressions for a variety of data types using the "simpleeval" library.
Description
Efficiency Nodes v1.62 Updates
UPDATES/CHANGES:
LoRA stacking is now supported for the 'Efficienct Loader' and XY Plot via the 'LoRA Stacker' node.
Massive code refactoring done to accomodate the new LoRA stacking changes.
BUG FIXES:
Fixed a bug where the XY Plot would not display the correct results when one XY type was "Sampler" and the other "Scheduler".
Fixed issues with XY Plot printouts.
FOR PEOPLE HAVING ERRORS:
The 'Efficient Loader' node's outputs have been changed. This means 'Efficient Loaders' in existing workflows must be manually replaced with a new ones from your node library.
FAQ
Comments (16)
If I get the latest, do I need any of the other versions, or does it contain the rest?
Get only latest. There is honestly no reason for me to keep the older ones except for archiving sake.
When sdxl support?
Hello, I tried to install and use it but get the error:
Error occurred when executing KSampler (Efficient): 'pooled_output' File "C:\Users\GABSU\comfyui\ComfyUI_windows_portable\ComfyUI\execution.py", line 144, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "C:\Users\GABSU\comfyui\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "C:\Users\GABSU\comfyui\ComfyUI_windows_portable\ComfyUI\execution.py", line 67, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "C:\Users\GABSU\comfyui\ComfyUI_windows_portable\ComfyUI\custom_nodes\efficiency-nodes-comfyui\efficiency_nodes.py", line 369, in sample samples
Have any ideas of what this could be? Thank you so much :)
Same, I think it doesn't work with SDXL checkpoint
@jospawn587 Yes, indeed. I tried using it with 1.5 and worked fine
waiting for SDXL support!
me too! the current sdxl workflows look like a fucking train wreck.....
Working on it
I want to use xy plot, but I get that error. please! help!
!!! Exception during processing !!!
Traceback (most recent call last):
File "C:\Users\laga4\.conda\envs\comfyui\ComfyUI\execution.py", line 193, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "C:\Users\laga4\.conda\envs\comfyui\ComfyUI\execution.py", line 126, in get_output_data
return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
File "C:\Users\laga4\.conda\envs\comfyui\ComfyUI\execution.py", line 119, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "C:\Users\laga4\.conda\envs\comfyui\ComfyUI\custom_nodes/efficiency-nodes-comfyui\efficiency_nodes.py", line 99, in efficientloader
model, clip, vae = load_checkpoint(ckpt_name, my_unique_id, cache=ckpt_cache, cache_overwrite=True)
File "C:\Users\laga4\.conda\envs\comfyui\ComfyUI\custom_nodes\efficiency-nodes-comfyui\tsc_utils.py", line 179, in load_checkpoint
out = load_checkpoint_guess_config_tsc(ckpt_path, output_vae, output_clip=True,
File "C:\Users\laga4\.conda\envs\comfyui\ComfyUI\custom_nodes\efficiency-nodes-comfyui\tsc_sd.py", line 274, in load_checkpoint_guess_config_tsc
parameters = calculate_parameters(sd, "model.diffusion_model.")
NameError: name 'calculate_parameters' is not defined
I'm using 1.5v that's not up to date, so it works fine.
How to disable the clip skip parameter in the efficient loader?
It demands the input to be at least -1, but this alters the result already compared to no clip skip
clip skip -1 is the same thing as no clip skip. if you think you had no clip skip before, you may have had a default clip skip of -2 (same thing as clip skip 2 in a1111). try setting to -2.
@pentimusoctem810 I tryied both, the result is still different from the regular loader, see here: https://imgur.com/a/SIRkqx4
@kymm Maybe remove the Lora from both of them and try? Just maybe....
Came across your comment because im having the exact same thing happening when im testing as well, everything identical yet the results vary dramatically even more so with people, did you end up ever figuring out a workaround or just went to using one or the other for consistency? thanks!