Hello, creators and builders,
While this week didn’t bring new model releases or research papers from our lab, it was anything but quiet.
In fact, it was a brilliant showcase of community ingenuity — developers building, refining, and reimagining what’s possible with AIGC ecosystem.
Let’s celebrate what you built this week.
👉 Subscribe to The Tongyi Weekly and never miss a release:
Subscribe Now → https://www.linkedin.com/build-relation/newsletter-follow?entityUrn=7392460924453945345
✨ Community Spotlights
Powerful ControlNet Union: Qwen-Image-2512-Fun-Controlnet-Union from alibaba-pai
A breakthrough in control: this new ControlNet Union from alibaba-pai integrates Canny, HED, Depth, Pose, MLSD, Scribble, and Inpainting into a single model — all built on 5 layer blocks of Qwen-Image-2512. No more juggling models. Just one pipeline, infinite control.
👉 Try it here
Unblur-Upscale: Qwen-Image-Edit-2511-Unblur-Upscale from prithivMLmods
This new adapter for Qwen-Image-Edit-2511 is designed to unblur and upscale images to high resolution. The model enhances image clarity by reducing blur, restoring fine details, and improving overall sharpness while preserving natural textures and realistic colors.
A masterclass in efficiency and visual fidelity.
👉 Try it here
Smater Control: Z-Image-Turbo-Fun-Controlnet-Union-2.1-2601 from Alibaba-PAI
Alibaba-pai just updated Z-Image-Turbo-Fun-Controlnet-Union-2.1-2601. The team has fixed the inference speed bug and significantly improved robustness by restructuring the dataset with multi-resolution control images (512~1536).
Highlight: A new 8-step distilled version is now available. It solves the blurriness issue found in previous tests and finally restores the model's true acceleration capabilities.
👉 Try it here
Image Rotation Restoration: Qwen-Image-Edit-2511-Gaussian-Splash from dx8152
Generate a PLY point cloud --> Adjust angle in editor -->Refine with Qwen-Image-Edit-2511 Gaussian Splash LoRA.
It accurately restores complex perspective shifts.
As shown in the demo, it handles 3D rotation and can even restore high-def details from blurry close-ups. Within a 45° range, the consistency is unmatched.
👉 Try it here
Anime Sketch Extractor: QwenImageEdit_LoRA from yeq6x
Interested in seeing the clean line art behind your favorite anime characters? This new LoRA for Qwen-Image-Edit-2509 effectively extracts sketches from existing images.
It’s a handy tool for anyone wanting to study character structure.
A must-have for animators and illustrators.
👉 Try it here
Sampling Optimized for Z-Image: ComfyUI-LG_SamplingUtils from LAOGOU-666
ComfyUI-LG_SamplingUtils is a comprehensive toolset designed for ComfyUI by LAOGOU-666, providing a series of practical sampling nodes that make operations more intuitive and convenient. This extension focuses on advanced sampling techniques, particularly optimized for Flow Matching models like ZImage and Lumina2.
👉 Try it here
🔥 Events & Challenges
New Challenge: #starringwithwan Is LIVE!
The #starringwithwan Challenge is officially open! Use the “Starring” feature in Wan App to create a video where you share the screen with our lead AI character Rowan and Ewan.🏆 Top 20 creators win 1-month Premium Membership (redeemable code)
🗓️ Deadline: January 28, 2026
👀 How to Enter
- Download the Wan App: https://wan.video/#wan-app?cref=mkt&cinfo=twitter
- Use the “Starring” feature to create a video with our lead character
- Post your video on X with hashtags #starringwithwan & #wanapp
📬 Want More? Stay Updated.
Every week, we bring you:
- New model releases & upgrades
- AI research breakthroughs
- Open-source tools you can use today
- Community highlights that inspire
👉 Subscribe to The Tongyi Weekly and never miss a release.
Subscribe Now → https://www.linkedin.com/build-relation/newsletter-follow?entityUrn=7392460924453945345
Thank you for being part of this journey.
Tongyi Lab is a research institution under Alibaba Group dedicated to artificial intelligence and foundation models, focusing on the research, development, and innovative applications of AI models across diverse domains. Its research spans large language models (LLMs), multimodal understanding and generation, visual AIGC, speech technologies, and more.
Top comments (0)