Get Stable Diffusion in your browser and access the world's most powerful AI art generation š
Joined May 2023
- Tweets2.3K
- Following226
- Followers1.3K
- Likes2.2K
ThinkDiffusion retweeted
NeedĀ moreĀ controlĀ overĀ yourĀ Wan2.2Ā videos?Ā
PrincepainterĀ hasĀ theĀ answer!Ā TheĀ ComfyUI-PainterI2VĀ nodeĀ isĀ builtĀ toĀ makeĀ yourĀ cameraĀ promptsĀ strongerĀ andĀ moreĀ responsive,Ā whileĀ alsoĀ boostingĀ overallĀ motionĀ byĀ 15-50%. SuchĀ aĀ fantasticĀ andĀ practicalĀ toolĀ fromĀ theĀ community @ComfyUI !Ā
GetĀ itĀ here:Ā github.com/princepainter/Comā¦
ThinkDiffusion retweeted
EAT IT ā a kinetic experiment in ComfyUI -- @VisualFrisson automated image sequences that react to the beat, generating rhythmic motion directly inside ComfyUI.
From folders to flow ā pure procedural animation with total control.
Weāll dive into the workflow live this Friday!
ThinkDiffusion retweeted
When it is time to go back to work...
Made with Wan ATI in @ComfyUI using @Machinedelusion's node
ThinkDiffusion retweeted
Our community trip to Tokyo in November 2024 was one of the most unreal experiences of my life, so I made it even crazier.
Donāt watch this twice. Iāve been dizzy since yesterday, and that's exactly how I feel about that memory.
Made in @ComfyUI
ThinkDiffusion retweeted
š¹ļø Wan ATI control with the new Path Animation Editor by @Machinedelusion in @ComfyUI
So. Much. Fun. šš
ThinkDiffusion retweeted
šļø Get Comfy With Comfy - Wan 2.2 Animate
In just 3 minutes, weāll cover a powerful model for character animation and movement inside ComfyUI.
This short tutorial will give you the essentials so you can start experimenting with character-driven AI animation right away.
ThinkDiffusion retweeted
š„ Qwen-Image-Edit-2509 IS LIVE ā and itās a GAME CHANGER. š„
We didnāt just upgrade it. We rebuilt it for creators, designers, and AI tinkerers who demand pixel-perfect control.
ā
Multi-Image Editing? YES.
Drag in āperson + productā or āperson + sceneā ā it blends them like magic. No more Franken-images.
ā
Single-Image? Rock-Solid Consistency.
ā¢ š¤ Faces stay you ā through poses, filters, and wild styles.
ā¢ šļø Products keep their identity ā ideal for ads & posters.
ā¢ āļø Text? Edit everything: content, font, color, even material texture.
ā
ControlNet Built-In.
Depth. Edges. Keypoints. Plug & play precision.
āØ Blog: qwen.ai/blog?id=7a90090115eeā¦
š¬ QwenChat: chat.qwen.ai/?inputFeature=iā¦
š GitHub: github.com/QwenLM/Qwen-Image
š¤ HuggingFace: huggingface.co/Qwen/Qwen-Imaā¦
š§© ModelScope: modelscope.cn/models/Qwen/Qwā¦
ThinkDiffusion retweeted
Wan 2.2 Animate is CRAZY and it actually excels at 3 things from my tests:
1. Lip syncing (so far the best open source I have seen, beating Runway Act2)
2. Consistent lighting & shadows with color tone replication when you swap a character
3. It keeps the replacement character aligned with realistic body dynamics even beyond the face. It is great for full body replacement
ThinkDiffusion retweeted
Qwen-Image InstantX Inpainting ControlNet is natively supported in ComfyUI
Professional inpainting & image editing capabilities now available!
š¹ Object replacement
š¹ Text modification
š¹ Background changes
š¹ Outpainting
Perfect for precise image edits with Qwen-Image's power. Ready to use in ComfyUI workflows!
ThinkDiffusion retweeted
Stable Audio 2.5 in ComfyUI ā the first audio generation model built for enterprise-grade sound production š¶
āØ Highlights:
Generate 3-min tracks in 2s
Multi-part compositions
Audio inpainting / extensions
Fully licensed dataset for commercial use
š Update ComfyUI ā search āStable Audio APIā
ThinkDiffusion retweeted
Been dialing my inpaint workflow for getting any subject into an existing shot
#comfyui @Alibaba_Wan