Wan 3.0 AI Video Generator
Create 2 to 30 second videos with Wan 3.0 on HeyMarmot. Combine text with image, video and audio references, generate synchronized sound, and export at up to 1080P.
Optional: add up to 10 images, 5 videos and 5 audio tracks, with no more than 20 reference files in total.
Up to 30 seconds
Build longer, coherent shots with native audio and precise output controls, without stitching together a stack of short clips.
Multimodal direction
One finished sequence
Adaptive ratio, up to 1080P
Wan 3.0 and the best AI video models in one studio
Start with Wan 3.0 Prime for fast high quality generation, switch to Wan 3.0 for a lower credit cost, or compare the same idea with other leading models.
Direct the whole sequence, not just the first frame
Wan 3.0 keeps subjects, motion and sound working together across clips as long as 30 seconds. Use multimodal references to communicate the look, movement and audio texture in one generation.
Create a Wan 3.0 video in three steps
Move from a creative brief to a finished clip without leaving the page.
Describe and reference
Write the scene and add any images, clips or audio that should guide the generation.
Choose Wan and settings
Select Wan 3.0 Prime or Wan 3.0, then set duration, aspect ratio, resolution and audio.
Generate and refine
Review the result, download it, or adjust the direction and create another version in the same conversation.
Choose the right model for every shot
Use Wan 3.0 Prime, Wan 3.0, Hailuo H3, Seedance 2.5, Veo 3.1 and more without changing tools or managing separate subscriptions.
Wan 3.0 Prime
Available in the HeyMarmot video workspace.
Wan 3.0
Available in the HeyMarmot video workspace.
Hailuo H3
Available in the HeyMarmot video workspace.
Seedance 2.5
Available in the HeyMarmot video workspace.
Veo 3.1
Available in the HeyMarmot video workspace.
Sora 2
Available in the HeyMarmot video workspace.
Kling 3.0
Available in the HeyMarmot video workspace.
What creators make with Wan 3.0
A longer generation window and multimodal direction make Wan 3.0 useful from story development to production ready social content.
Narrative sequences
Create longer establishing shots, character beats and camera moves with less need to join separate clips.
Social video
Generate vertical stories and product explainers with pacing, dialogue and sound designed together.
Product campaigns
Guide materials, composition and motion with product images and reference footage, then export at 1080P.
Character and performance
Use image, video and audio references to shape identity, movement and vocal atmosphere in the same shot.
A complete Wan 3.0 workflow
HeyMarmot keeps model controls, references, results and iteration history together so you can move from test to final render quickly.
Multimodal prompting
Combine a detailed prompt with up to 10 images, 5 videos and 5 audio references.
Frame guided motion
Use frame inputs in the workspace when the opening or closing composition needs precise control.
Synchronized audio
Generate sound with the picture, or turn audio generation off when you plan to finish sound separately.
480P to 1080P
Choose a fast draft resolution or render the final result in full HD.
Reusable asset library
Keep references and generated outputs ready for later variations and new conversations.
Conversation history
Prompts, settings, source files and every Wan result stay connected in one thread.
Wan 3.0 FAQ
Practical answers about creating Wan 3.0 video on HeyMarmot.
What is Wan 3.0?
Wan 3.0 is an Alibaba video generation model for text to video and multimodal reference driven creation. HeyMarmot offers both Wan 3.0 Video and the faster Wan 3.0 Video Prime.
What inputs can I use with Wan 3.0?
You can generate from text alone or add image, video and audio references. The reference mode accepts up to 10 images, 5 videos and 5 audio tracks, with no more than 20 files in total.
How long can Wan 3.0 videos be?
You can choose any whole second from 2 to 30 seconds for a single generation.
Which resolutions and aspect ratios are supported?
Wan 3.0 supports 480P, 720P and 1080P. Aspect ratios include adaptive, 16:9, 4:3, 1:1, 3:4 and 9:16.
Does Wan 3.0 generate audio?
Yes. Audio generation is available and can be turned on or off before submitting the video.
What is the difference between Wan 3.0 and Wan 3.0 Prime?
Both versions expose the same duration, resolution, aspect ratio and reference controls on HeyMarmot. Prime is positioned for faster high quality generation and uses more credits per second than the standard model.
Keep exploring AI video
Browse examples, compare workflows or check credit pricing before your next generation.