Wan
@Alibaba_Wan
Followers
25K
Following
100
Media
89
Statuses
375
Video & Image Generation Model from @Ali_TongyiLab
Joined February 2025
Today, we're officially launching Wan2.5-Preview! It's set to reshape the future of visual generation with a new architecture and powerful features. • Architectural Features: Native Multimodality, Deep Alignment ∘ Native Multimodal Architecture: Adopts a new, unified framework
171
236
2K
Made a 20-second 480p AI animation in just 15 minutes on a 12GB GPU. WAN 2.2 i2v is actually insane 🤯🔥 Upscled it to 1080p using https://t.co/TgfUaOIgXB For workflow : https://t.co/qPgMlJRpqE
#wan2 #aivideos #Pokemon
2
2
15
nice try
I am trying to mix Motion Graphics and AI. This is a mixture of After Effects and AI model Wan2.2. At the same time, I think its a great tool to animate style-frames that are very difficult to execute. #comfyui #aftereffects #ai #aivideo #motiongraphics
0
1
12
😍 So nice！
1
2
15
Great job！
Guess what? WAN again. EgoEdit: a real-time streaming editor; - preserves hand-object interactions; - distilled Wan 2.1; 38.1 FPS with 855ms latency on a single H100. Primary usage area - AR and Mixed Reality. https://t.co/gUAdEvSjvj
0
0
11
Alibaba doesn't stop cooking. They just dropped a REALTIME, infinite length video generator. Based on Wan, 20 fps, with dialogue https://t.co/vbEmdyl500
28
116
1K
Our NeurIPS 2025 paper Wan-Move is now #1 on Hugging Face Daily Papers 🎉 TL;DR: Bring Wan Image-to-Video to SOTA fine-grained, point-level motion control! 🎯 5s 480p high-quality motion control, on par with commercial systems 🧩 A new latent trajectory guidance for I2V models
0
3
12
🌟Amazing work—huge congrats to the authors! Check it out!
Our NeurIPS 2025 paper Wan-Move is now #1 on Hugging Face Daily Papers 🎉 TL;DR: Bring Wan Image-to-Video to SOTA fine-grained, point-level motion control! 🎯 5s 480p high-quality motion control, on par with commercial systems 🧩 A new latent trajectory guidance for I2V models
1
2
38
WanMuse+ Season 2 Global Creative Competition「Hearbeat」– Winners Announced With heartbeats as our measure, we dive into the depths of creativity; With the world as our mirror, we reflect the brilliance of inspiration. After multiple rounds of evaluation by the WanMuse Jury,
4
0
28
Now Live | WanMuse + 「Heartbeat」Professional Category Mashup 30 award-winning creations have arrived—this time, AI isn’t just generating… it’s genuinely moving. 🎬 Frames rise with every heartbeat, light shapes pure emotion, and every second pulses with the harmony of tech and
2
5
42
🎬 The "Heartbeat" Professional Category Shortlist Is Here! Check It Out → Wan Muse Season 2 “Heartbeat” Event has officially concluded. We received over 10,000 submissions from creators around the world—diverse in topics and rich in style—fully showcasing the vibrant potential
6
1
21
This is exciting news from @lmsysorg ! SGLang Diffusion is here, bringing great acceleration to diffusion models. We love to see this kind of ecosystem support, and we're happy to see that it includes support for Wan-series. Congrats to the SGLang Diffusion Team! 🔗:
lmsys.org
We are excited to introduce SGLang Diffusion, which brings SGLang's state-of-the-art performance to accelerate image and video generation for diffusion mo...
0
10
50
What a fantastic milestone! Thanks for the shoutout, @arena . The entire @Alibaba_Wan team is absolutely thrilled. A huge amount of hard work and dedication went into the Wan 2.5 Preview, and seeing this kind of reception from the community is incredibly rewarding for all our
🎬 🖼️ Video & Image Leaderboard Update Wan 2.5 preview models i2v and t2i have landed in the Top 5 on the Image-to-Video and Text-to-Image leaderboards. 🔹Wan 2.5-i2v-preview ranks #3 on Image-to-Video 🔹Wan 2.5-t2i-preview ranks #5 on Text-to-Image Congrats to the
7
13
81
Proud to see Wan among the leading image & video models powering ElevenLabs’ new unified creative platform. Excited to help creators and brands bring ideas to life with higher quality, more speed, and more control. Try it now on ElevenLabs Image & Video.
Introducing ElevenLabs Image & Video - the best audio, image and video models now in one platform. Generate with leading models like Veo, Sora, Kling, Wan and Seedance, then enhance with the highest quality voices, music, and sound effects.
2
7
101
This is a seriously impressive workflow. 🤯 @DimScariot is piping Blender 3D trajectories right into Wan (via @ComfyUI) to create this super-controlled AI-generated motion. That custom Path Animator is doing amazing work. Fantastic stuff, and a great nod to @MIZNOM's work!
Blender 3D trajectories → JSON → injected into Wan (ComfyUI) with a custom Path Animator to drive AI-generated motion. Inspired by the amazing work of @MIZNOM
2
17
161
Wan 2.2 Animate → Character Replacement on Comfy Cloud Swap characters in a scene automatically with our new one-node workflow. No complex setups — just pick your character and go. Create cinematic replacements in seconds. 🚀
10
40
358
Need more control over your Wan2.2 videos? Princepainter has the answer! The ComfyUI-PainterI2V node is built to make your camera prompts stronger and more responsive, while also boosting overall motion by 15-50%. Such a fantastic and practical tool from the community @ComfyUI !
6
23
168
Love the incredible image quality of Wan 2.1 but wish your shots had more dynamic movement?Well, @OdinLovis has the perfect solution! This Motion LoRA was trained over 40 iterations to add one specific, high-quality effect: a "Push-in camera" motion. Just add the trigger word
5
17
142