ComfyUI Wiki
@ComfyUIWiki
Followers
750
Following
57
Media
24
Statuses
73
Currently maintaining the official ComfyUI documentation, template content, etc.
Joined September 2024
Is there anyone who still doesn't know Soham Parekh?🤣.
We worked with Soham Parekh for 3 weeks in June. He seemed extremely cracked in the technical interview, and said all the right things. That he loved open source, and ComfyUI. Then things started to fall apart.
0
0
3
Lots of desktop users may not know how to activate the desktop Python environment, so today I added a new section about it to the desktop docs. For more details:
2
4
30
I finished the OmniGen2 templates and docs this weekend. The PR hasn't been merged yet, but since OmniGen2 was supported before the Flux.1 Kontext Dev release, you can get the workflows from our docs:
OmniGen2 is natively supported in ComfyUI!.Generate, Edit , Compose. All unified. The 7B-parameter unified multimodal model unifies all functions：. - Text-to-Image Generation.- Instruction-guided Editing.- Multi-Image Composition.- Text in Images.- Visual Understanding.-
0
11
71
Updated Flux.1 Kontext Dev ComfyUI tutorials on my website today. Hope it helps:
2
40
198
RT @c__byrne: does anyone know someone who understands:.- vue/react compilation internals, vdom.- browser rendering pipeline & layerization….
0
4
0
Creating this video took me a lot of time. Hope you guys like it.
🎉Generate. Edit. Repeat. All locally. Flux.1 Kontext [dev] is now open source and fully supported in ComfyUI!. Everything below is now enabled:. 👉Multi-Step Editing with Context.👉Character Consistency.👉Local Editing.👉Style Reference.👉Object / Background Removal.👉Multiple
11
15
153
ComfyUI now has native support for Omnigen2. Follow this PR for more information:.
0
3
13
You can find the Flux Kontext API nodes workflows in the template under the Image API section
0
0
1
WOW! No complicated workflow, just upload your image and add prompts.🕺.
1
0
11
I’ve added lots of videos and images to the docs to help you better understand what each setting does.
0
0
1
I completed the ComfyUI settings docs yesterday! You can now check the settings docs here: If you spot any errors, feel free to point them out here:
3
1
17
I'm trying to add more examples to make the node parameters easier for you to understand, and I hope it helps.
0
0
1
The first doc version is from my site. Many of them were updated during my early days with ComfyUI. I’m still updating the node docs, but full completion will take time. Report errors via issues or PRs in this repo:
🎉 The built-in node help menu is now available to everyone. 🔁 Just update your ComfyUI to the latest version and make sure the frontend is at least v1.22.2
2
1
15
It now supports English! Thanks for the comment from @wong216224.
📢 ByteDance released its ComfyUI custom node "Comfyui-lumi-batcher". ✨ It allows you to control ANY parameters you want to compare, more powerful than the XYZ plot in WebUI. You can find it in ComfyUI Manager. It only has a Chinese interface now. 😢
0
0
4
Let us know your interested topic via comment, and I will create more tutorials like this. BTW:in this video, I'm using the Cosmos Predict2 2B. It's quite small and performing well.
📺 Watch this video to learn batch image gen with varied prompts using ComfyUI Inspire Pack. 🛠️ 2 methods in the video: 1-click (wait full run) or instant view (requires simple counting). ♥️ Like if you enjoy such tutorials and comment with your topic preferences!
2
0
7
💡A quick tip about workflow management:. 1⃣When saving a workflow, you can use "folder name/" to create a folder. 2⃣And if the workflow is loaded from a folder, it will create a subfolder.
1
0
14
RT @PurzBeats: We're going to try out the Nvidia Cosmos Predict-2 Models today live on stream!.
0
2
0
Have fun with Cosmos-Predict2 in ComfyUI.🕺🕺🕺
🎉 ComfyUI now natively supports NVIDIA’s Cosmos-Predict2 model family!. Cosmos-Predict2 brings high-fidelity, physics-aware Image generation and Video2World (Image-to-Video) generation. Another reality inside ComfyUI!
0
1
20