Wan 3.0 is currently presented as a beta workflow
Public Wan 3.0 specifications are still developing. Ezier therefore describes only the controls available in the current product instead of repeating unverified third-party parameter claims.
Create 30-Second Videos with Wan 3.0
Use Wan 3.0 on Ezier to create 2-30-second videos from text or one start image, choose 480P, 720P, or 1080P, and generate audio with the current beta controls.
Build a complete 30-second sequence with Wan 3.0
Use the longer beta runtime for a scene with a beginning, development, and finish. Give every action a clear order, keep the camera path readable, and place sound cues where they belong in the picture.
Prompt example
Create a 5-second cinematic farm-to-table produce montage in warm natural daylight. Open with an extreme macro shot of two hands gently twisting a ripe red tomato from a healthy green vine, preserving the water droplets, fine stem hairs, and soft garden background. Use a clean match cut into a bright kitchen where the same hands guide a stainless-steel chef knife through a fresh cucumber on a wooden board, ending on a precise stack of thin round slices. Keep the hand appearance, sunlight direction, fresh-produce texture, and shallow depth of field consistent across the transition. Use crisp tactile detail, slow controlled motion, and realistic contact between fingers, fruit, knife, and board. Add soft leaves, a natural stem snap, a clean knife rhythm, and quiet kitchen ambience. No text, logos, deformed fingers, floating produce, unsafe knife handling, crushed ingredients, or abrupt color changes.
How to use Wan 3.0 in Ezier
- 01
Select Wan 3.0
Confirm that Wan 3.0 is active in the model selector before choosing resolution, duration, ratio, or generated audio.
- 02
Choose text or one start image
Begin from a prompt, or upload one image when the opening composition, subject design, product geometry, or lighting must be anchored.
- 03
Write the sequence in time order
Describe the opening, action beats, camera movement, sound sources, continuity rules, and final frame in the order viewers should experience them.
- 04
Set the output and generate
Choose 2-30 seconds, 480P, 720P, or 1080P, a supported aspect ratio, seed when useful, and generated audio. Review the credit estimate, then generate.
Direct motion, pacing, and sound with Wan 3.0
A longer clip needs more than a longer prompt. Define the sequence in chronological beats, protect the opening composition when using an image, and connect every sound to a visible source or space.
Use the 30-second window for a real progression
The current Ezier beta control extends to 30 seconds. Spend that time on connected beats—a setup, a change, and a finish—instead of stretching one action without development.
Protect the opening with one start image
Use image-to-video when exact shape, color, framing, or lighting matters at the start. Describe what should move, what must remain unchanged, and where the final frame should settle.
Tie generated sound to visible events
Name the sound source, timing, distance, and environment. Visible causes such as footsteps, machinery, water, doors, or instruments give the audio plan a structure that can be reviewed with the motion.
Wan 3.0 beta settings available in Ezier
These are the controls currently exposed by Ezier for Wan 3.0. Because the model is in beta and public specifications are still developing, use the workbench as the source of truth for each generation.
- 2-30s
- Duration in Ezier betaUse a short clip for one controlled motion or more time for connected beats with a clear beginning and finish.
- 480P-1080P
- Resolution rangeChoose 480P for low-cost direction tests, 720P for balanced iteration, or 1080P when the selected idea needs a higher-resolution review.
- 6 + Auto
- Aspect ratio choicesUse Auto or choose 21:9, 16:9, 4:3, 1:1, 3:4, or 9:16 for the intended placement.
Test direction before spending on 1080P
Use 480P or 720P to confirm the action order, framing, and camera path. Move to 1080P after the sequence itself is worth keeping.
Give every beat one clear job
Each beat should introduce information, change the action, or deliver the finish. Remove repeated shots that do not move the sequence forward.
Keep start-image constraints specific
Name the exact geometry, colors, lighting, and framing to preserve. Then describe motion separately so the model is not asked to redesign and animate at once.
Review sound and picture on the same pass
Check whether each effect starts with the visible action, stays in the correct space, and ends naturally instead of evaluating the soundtrack as a separate layer.
What the current Wan 3.0 workflow supports
Ezier keeps the beta workflow focused on two clear starting points: describe a scene from text or animate one opening image, then choose the output settings you can review now.
Text-to-video and image-to-video are available now
Start from a written brief or attach one opening image. The current Ezier upload schema does not expose video, audio, document, slide, or multi-image reference uploads for Wan 3.0.
The current output ceiling is 30 seconds at 1080P
Ezier exposes durations from 2 to 30 seconds and resolution choices of 480P, 720P, and 1080P. The selected quality and duration affect the displayed credit estimate.
Generated audio is part of the current controls
Use the Generate audio switch when effects, ambience, or music matter to the result, and write those sounds into the same chronological plan as the visuals.
Wan 3.0 in Ezier
Start with one sequence you can review from beginning to end
Write the scene in order, add a start image only when the opening composition matters, choose duration and resolution, then generate and review motion, continuity, framing, and sound together.
Wan 3.0 FAQ
Answers about beta access, duration, resolution, image-to-video, generated audio, aspect ratios, prompts, and the controls currently available in Ezier.
Keep testing
More routes from Wan 3.0
Compare a nearby model, finish the clip with tools, or rebuild the source for another render.
AI models
Compare models
Choose a better fit before you generate.
AI video tools
Turn images into motion
Animate a still or create a clip.
AI image tools
Create or edit images
Make, clean up, or upscale stills.
MiniMax H3
2K multimodal video
Use for 2K clips with multi-image direction and generated audio.
Seedance 2.5
Longer reference-led video
Use for longer scenes with richer reference direction.
Veo 3.1
Cinematic control
Use for cinematic shots and camera moves.
Kling 3.0
Action and motion
Use for action, motion, and story beats.
Omni Flash
Conversational editing
Use for short video generation and conversational edits.
Hailuo AI
Physical motion
Use for fabric, liquid, and physical motion.
AI models
Compare models
Choose a better fit before you generate.
AI video tools
Turn images into motion
Animate a still or create a clip.
AI image tools
Create or edit images
Make, clean up, or upscale stills.
MiniMax H3
2K multimodal video
Use for 2K clips with multi-image direction and generated audio.