What modes does Wan 3.0 support?

Text-to-video, image-to-video (with optional end frame), and reference-to-video with up to 10 reference images plus optional reference audio or video.

What durations and resolutions are available?

Wan 3.0 supports 2 to 30 second clips at 480p, 720p, and 1080p, with aspect ratios including 16:9, 9:16, 1:1, 3:4, and 4:3.

How much does Wan 3.0 cost?

Pricing is 10 gold per second at 480p, 20 gold per second at 720p, and 40 gold per second at 1080p. Generated audio and thinking mode do not add cost.

Is Wan 3.0 subscriber-only?

Yes. Wan 3.0 is Premium-only, with one free trial use for eligible free users.

Does thinking mode cost extra?

No. Thinking Mode is free. It can help with complex multi-subject or multi-beat prompts.

Wan 3.0 Video Generator

Alibaba Wan 3.0 AI video with text-to-video, image-to-video, multi-reference guidance, thinking mode, and 480p to 1080p output.

Keep building

Build a character around the artStat block, backstory and portrait in one pass.Open the full studioEvery model, every content type, one workspace.Bring a map to lifeAnimated battlemaps for your next session.

Key Features

Text-to-Video, Image-to-Video, and Reference-to-Video under one model name

Up to 10 reference images for guided multi-subject clips

Optional end-frame control on image-to-video for deliberate final shots

Generated audio toggle and optional reference audio for reference-to-video

Thinking Mode for deeper understanding of complex prompts

Flexible 2s to 30s durations at 480p, 720p, or 1080p

Wan 3.0 Specifications

Model NameWan 3.0

Input ModesText-to-Video, Image-to-Video, Reference-to-Video

Image Inputs1 start frame (i2v) or up to 10 reference images (ref2v)

Resolutions480p, 720p, 1080p

Durations2s to 30s

Aspect Ratios16:9, 9:16, 1:1, 3:4, 4:3

Advanced ControlsGenerate Audio, Thinking Mode, End Frame

Pricing10 gold/s at 480p; 20 gold/s at 720p; 40 gold/s at 1080p

Best Uses for Wan 3.0

Cinematic fantasy trailers

Build short campaign cutscenes and hero reveals with flexible duration and resolution tiers.

Portrait-guided character clips

Start from a character image, optionally lock an end frame, and animate a controlled reveal or action beat.

Multi-reference consistency

Supply several reference images when you need outfits, faces, or locations to stay coherent across a clip.

Wan 3.0 vs Wan 2.7

Wan 3.0

Text-to-video, image-to-video, and reference-to-video under one name
480p, 720p, and 1080p with 2 to 30 second clips
Up to 10 reference images plus optional reference video or audio
Thinking Mode and Generate Audio at no extra gold

Wan 2.7

Premium 720p and 1080p shorts with prompt expansion
End-frame control and audio upload on image-to-video
Negative prompts for cleanup and style control
2 to 15 second durations without a 480p tier

More about Wan 3.0 on CharGen

Generate Wan 3.0 clips from the Video Studio picker, then send results into Gallery or Media Center. Gold is charged server-side from resolution and duration, and the UI estimate matches that formula.