Home
Creators
Search Recent Missing Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

You can install NO8D-Controls in the manager (Patreon)

Published:
2026-07-09 03:17:58
Imported:
2026-10
Tags:

Content

NO8D-Controls has been registered in Comfy and will be officially launched today! You can consider this version as the official release!

This version has rewritten the inpainting node, and now it's called Generate.

The biggest change is that it has increased the generation speed by 100%(which may vary slightly depending on different graphics cards); at the same time, it supports wan2.1_vae_upscale2x, doubling the quality of generated images; negative will default to "condition nullification" when there is no input, meaning you no longer need to connect the negative node; local redrawing has added an eraser function, and the feathering effect can be directly seen on the canvas, making your operations more intuitive.

The stack node has added a setting to save trigger words. You no longer need to remember its trigger words in the future. After setting it up, simply connect the output of the trigger word to the fixed prompt of the prompt node.

The logic of the prompt node has also been optimized. Now, the prompt node no longer requires text input; you can directly fill in the content in the input text, and it will automatically complete the expansion of the prompt. When there is both text input and image input, the two can work together, with text expansion having a higher priority. Additionally, commonly used shot control has been newly added.

The Images node has added automated judgment. If you select several images here and click Run, the node will output the same number of images. If you don't select any and click Run, the node will output all images, which is very suitable for batch processing tasks. It also supports mouse + keyboard selection operations, allowing you to select by dragging a box, multi-select, invert selection, and also supports dragging images to adjust the order. It supports directly dragging images from other folders. You can delete images with one click in the window (without deleting the original images).

https://github.com/no8d/ComfyUI-NO8D-controls

OK, the above is all the updates for this time. Now you can directly obtain "NO8D-Controls" by searching for it in comfyui-manager. "NO8D-Controls" is entirely designed to improve efficiency and enhance control. I hope you will like it.

🫡 May the AI-power be with you, see you soon !🫡

Disclaimer: My model or workflow is solely for learning and communication purposes and shall not be used for any commercial activities. All responsibilities for the generated content, including its legality, ethical implications, or potential misuse, are borne entirely by the user. By using this model or workflow, you agree to this principle. Please use it responsibly within the scope of local laws.

Files

Comments

Wylmquest

I have some comments. I want to include screenshots. How?

Wylmquest

[INFO] got prompt [ERROR] !!! Exception during processing !!! NO8D-Prompt: model is empty [ERROR] Traceback (most recent call last): File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\execution.py", line 542, in execute output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\execution.py", line 341, in get_output_data return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, v3_data=v3_data) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\custom_nodes\comfyui-lora-manager\py\metadata_collector\metadata_hook.py", line 171, in async_map_node_over_list_with_metadata results = await original_map_node_over_list( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<2 lines>... ) ^ File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\execution.py", line 315, in _async_map_node_over_list await process_inputs(input_dict, i) File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\execution.py", line 303, in process_inputs result = f(**inputs) File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\custom_nodes\no8d-control\prompt_plus.py", line 1072, in run self._generate_prompt_item( ~~~~~~~~~~~~~~~~~~~~~~~~~~^ index=index, ^^^^^^^^^^^^ ...<2 lines>... **common, ^^^^^^^^^ ) ^ File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\custom_nodes\no8d-control\prompt_plus.py", line 1001, in _generate_prompt_item raw = _chat_completion(api_base_url, api_key, model, messages, temperature, max_tokens, effective_seed, service_type) File "H:\ComfyUI_5\ComfyUI_windows_portable\ComfyUI\custom_nodes\no8d-control\prompt_plus.py", line 801, in _chat_completion raise ValueError("NO8D-Prompt: model is empty") ValueError: NO8D-Prompt: model is empty [INFO] Prompt executed in 0.03 seconds

Wylmquest

NOt now, but probably i could...

NO8D's AI-Workshop

The NO8D-prompt node requires API or LLM configuration to function properly

Wylmquest

oh, i see, very clever, and how i do that? any instructions available???

NO8D's AI-Workshop

You can find the NO8D settings in ComfyUI's configuration. If you have the ChatGPT API key, just enter it directly to start using it. If you use other AI services, fill in the URL and API key, then you can utilize the feature.

Runebinder

Just to make you aware that although it is in the Manager, I'm getting the security level alert and it won't install. I know how to fix it, just a heads up.

Runebinder

Ignore below. I did try clicking Validate API and nothing happened. So tried the god old click it multiple times and I've got my model list now. On the back of this, I went into settings, NO8D > API Manager > Add Service and selected Local LLM (Ollama) which I have set up with quite a few models, but I don't see any way to select an Ollama model. Tried saving but the node still gives the error for no model selected. How do I fix that and select a model?

Runebinder

I've set up Ollama for the prompt node with Gemma 4 12b for both text and vision models, but when I try and run the workflow I get am error: "NO8D-Prompt: Ollama returned empty content" (let me know if you need the full error) The model works fine in the Ollama nodes, am I missing something with the settings?

Wylmquest

Hello again, I need to control seed and its iteration and fix of the generation node, from outside the subgraph. Typical way of exposing widgets does not work for me. The padlock is not accessible otherwise than manually? Please explain if there is any way like with standard nodes. Thanks in advance.

NO8D's AI-Workshop

Sorry, I just saw your message. I'll check the code. Could you send me the complete error message?

Wylmquest

I am asking to be able to control the seed from outside the node and to be able to use your node inside subgraph with the seed and seed control visible as widgets on the subgraph. If you do not understand what I mean make some research. Standard nodes do not use padlocks. There are 4 options for seed: fix, increment, decrement, random. Seed can be generated inside the node or in any relevant seed generator connected to the node in the seed widget. That is standard comfyui node design. Your generation not is very good, the only thing you need to do is to standardize parameters connectors. Just follow comfyui template for that part. No fancy padlocks please. Thanks in advance.

Wylmquest

Ok i had no time to wait for you so i modified generate.js myself using ChatGPT. it works perfectly well now as standard comfyui node. If you are interested you can check your discord. i will send you the parches. you can use them in your next publication with acknowledgments. I also discovered that your generate node (and probably others too) is written accordingly to old V1 standard of Confyui org. So probably rewriting them in V3 standard can be your next project.

Runebinder

Just saw your response and run the workflow and of course it ran fine first time. Safe to say can file it away as an annoying temporary glitch.

Wylmquest

HI NO8D, i am finally very irritated that you insist on having not standard layout of the generate node. It is very, very annoying. Each time you make an update i have to fix this node myself to restore its standard look and functionality of widgets. I gave you on Discord detailed information on how to make this node standard. It is very easy and you should do that right now. for next update. If you don't know how, let me know, i will help you.

NO8D's AI-Workshop

Sorry mate, if you check Discord, you'll see I explained the original intention behind this node's design to you right away a long time ago. Therefore, its interface won't be changed in the short term. The good news is that its inpaint/outpaint functions have been initially built, so there will be no further changes to it anytime soon. You can modify it to your preferred layout without worrying about updates overriding your changes.