ComfyUI

Comfy Desktop is here - one app for every Comfy. We rebuilt the desktop experience from the ground up. Your existing workflows, custom nodes, models, and settings carry over untouched. Same name, completely new app. What's new: → Multiple ComfyUI instances — different custom nodes, different versions, all managed in one place → Automatic snapshots before every update and custom node change, with one-click rollback if something breaks  → Day-0 ComfyUI releases — the moment a new version or nightly drops, you can update immediately If you're on our older desktop app, you'll see an update prompt automatically.

Creators IG: instagram.com/seungho__yeo/ Creators IG: instagram.com/whatzuwant/

Creator seungho__yeo ( IG ) built an entire anime-inspired 2D motion sequence - storyboard to final render inside Comfy. seungho__yeo created this workflow while creator whatzuwant ( IG ) created the fully AI generated anime + sound design. pipeline: → Poster art → Storyboard frames → Transitions + animation → Edit + final render What usually takes multiple apps was built in one tool.

@JulienAIArt @krea_ai

@ComfyUI @krea_ai Ask and you shall receive!!!

Made these using Krea 2 Turbo locally in ComfyUI during our stream today. So far this model seems extremely capable and artistic. Have you tried it yet? @krea_ai @ComfyUI

Krea X Comfy: Founders Live x.com/i/broadcasts/1…

@MarcinKrajews18 @krea_ai

@ComfyUI @krea_ai Holy shit it's so good, how is this even possible. Small turbo model with solid world knowledge? Things are moving way too fast lately.

Krea 2 is back and this time the weights are OPEN. This open source release ships with two models designed to work together — @krea_ai 2 RAW and Krea 2 Turbo. RAW is for training, Turbo is for inference.

today, we release the open weights of Krea 2. welcome Krea 2 Raw and Krea 2 Turbo, an undistilled model from mid-training meant to be fine-tuned, and a fast distilled version with a wide aesthetic diversity. read the details below 👇

Krea x Comfy: Founders Live A special live conversation with our guests Victor Perez (CEO, Krea), Miguel Lara (Krea Team), and ComfyAnonymous (Co-Founder, Comfy Org), hosted by Purz & Julien. Today at 3pm PST!

Open Source again for the win. Go Get it!

Ojito con este nuevo modelo 👌

Getting Started 1. Update ComfyUI to the latest version 2. Search "Krea 2 Turbo" in the template library 3. Follow the note in the workflow to download models and run the workflow github.com/Comfy-Org/work…

Read the technical paper on Krea 2 krea.ai/blog/krea-2-te… Download the model weights github.com/krea-ai/krea-2

Creators IG: instagram.com/a01demort/

Creator a01demort ( IG ) built this hype film locally for basically $0. The pipeline: → Image Generation (ChatGPT) → Video Generation (LTX 2.3) → Lip Sync (LTX 2.3) → Motion & Compositing (ComfyUI Video Combine) → Final Edit (CapCut) Production cost: basically $0. ( His Chat GPT subscription was the only thing paid for ) Not bigger budgets. Better pipelines. This is the level ComfyUI creators are working at.

magnet:?xt=urn:btih:2a644d0279182a022d08dd395ea593cfcc218e12&dn=watering-hole.zip

Creators IG: instagram.com/seungho__yeo/

The hardest part of VFX used to be making it look like it belongs. That That bar just got cleared. What used to require tracked plates, manual roto, and a lighting team now starts with a ComfyUI workflow and an auto-mask. Creator seungho__yeo ran ComfyUI + Beeble AI Canvas to blend AI-generated objects directly into live video: → Objects generated in ComfyUI, dropped into real footage without breaking the scene → Auto-masking that holds the original depth, lighting, and spatial feel → Per-mask motion control — follow the camera or move independently → Generate → Mask → Composite → Done The dolphin composite alone shows what's possible: different mask regions, different motion behaviors, one coherent shot. This is what VFX looks like when the pipeline fits in a single creator's hands.

This ComfyUI workflow lets you draw lines and annotations to guide motion, define where actions should happen, and influence movement throughout a scene. An integrated LLM translates those annotations into prompt instructions, making it easier to direct motion visually instead of relying entirely on text prompts. The workflow works best with clear, intentional paths, while more complex routes with frequent direction changes can be harder to follow consistently. To try this workflow, link below 👇