Prompt Wan for image-to-video

Last updated: October 12, 2026

Wan 3.0 animates an approved still into a clip that follows the camera you ask for, with sound generated in the same pass. Pick it when the frame is signed off and you need to control the shot: a held idle, a cast on a locked camera, a slow push in on a hero.

Let the agent write it

You don't have to write any of this yourself. The Layer agent knows how to prompt each model it runs, from the same guidance this page is built on. Attach the still in chat, name the model and say what should happen, for example: Animate this with Wan 3.0: a slow fire-spell cast, fixed camera, 2 seconds at 480p, with a whoosh and crackling flame, no music. The agent writes the prompt, shows you its plan, and runs it when you approve. Use the rest of this guide when you want to write or adjust the prompt yourself, and see Write a brief for the agent for how to brief it.

When to use it

Use Wan when you have the frame and need it to move in a shot you can direct: character idles and ability casts, hero reveals with a push in, and ambient motion for menus and backgrounds. It keeps the character's design and style from the first frame, follows plain camera terms, and runs from 2 to 30 seconds, so you can find the motion on a short, low-resolution draft and render the length you need.

It needs the shot written out: from a one-line prompt it invents its own camera move and effects. To animate the same frame with another model, see Prompt Seedance for image-to-video. To generate a clip with no image at all, see Create a video.

Versions on Layer

Wan 3.0

The version this guide's examples were made with. Clips run 2 to 30 seconds at 480p, 720p or 1080p, with a first and a last frame, and native audio.

How to write the prompt

Camera, then the motion in order with its speed, then the style, the sound, and what must not appear:

Fixed camera, medium shot. The mage slowly raises the staff with both hands as the ember at its tip swells into a ball of fire, then thrusts it forward and the fireball shoots out of frame to the right; his cloak flaps back with the force. Stylised painterly game art. Sound: a rising whoosh and a crackling burst of flame, no background music. No dialogue, no subtitles, no text, no watermark.

Alibaba's formula for a first frame is motion plus camera movement: the image already sets who is in the shot, where, and how it looks. Putting the camera first fixes the shot before the action starts, and writing the action as a sequence tells the model what happens in what order. That prompt is the one in the clip above.

What works

Spell out the camera and the motion

We ran the prompt above against "The mage casts a powerful fire spell." on the same frame and seed. The written-out prompt held the camera still, kept the mage on-model and landed each beat in order: the staff rises, the ember swells, the thrust. The one-liner chose its own push in, wrapped the mage in a ring of fire and ended in a full-frame white flash, with nothing left of the character. Both stayed in one take; only the specific prompt gave a usable game asset.

Name one camera move

Wan follows camera terms as written. On the same frame, seed and action, "fixed camera" kept the ledge and the castle in place from the first frame to the last, and "slow push in toward the mage's face" closed from a full-length shot to the waist while the same cast played out. Terms from Alibaba's guide: fixed shot, push in, pull out, orbit, handheld follow, lateral dolly, crane, zoom, whip pan.

Speed in plain words

Control speed and force with words such as slowly, quickly, gently and extremely slow. Leave out counted, high-frequency actions such as "shakes his head three times a second": the model cannot time them.

Name the style, tone and mood

Close the visual part with the medium ("stylised painterly game art", "line illustration", "clay animation"), and add a colour tone ("low-key high-contrast lighting") or a mood ("mysterious and magical") when the shot needs one.

Sound in the prompt

With Generate audio on, Wan fills in sound from the action and materials it sees. Name the sounds you want ("a rising whoosh and a crackling burst of flame"), and write "no background music" or "no dialogue" to keep them out. For speech, write the speaker, a speaking verb and the line in quotes: the mage whispers: "Burn." About two short sentences fit a 5-second clip; longer speech gets rushed or cut.

Several shots in one clip

Write numbered shots with their times, each 2 to 5 seconds and end to end with no gaps: "Shot 1 [0-3s]: the mage raises the staff. Shot 2 [3-6s]: close-up on the ember as it ignites." Keep the character in the same place and costume across the cut.

A last frame for the ending

Add a Last frame when the ending pose or composition matters, such as a cast that must finish in the same stance as a sprite's idle. Wan fills in the motion between the two frames.

What to avoid

Expecting a negative prompt

Wan 3.0 has no negative prompt. End the prompt with a short list of what must not appear: "No dialogue, no subtitles, no text, no watermark. No character position changes, no costume changes." List only what you don't want, and don't repeat what the prompt already asks for.

Impossible motion

Movement that breaks the body's centre of gravity or joint range, and superhuman feats (one character hurling a much heavier one with one hand), come out broken. Keep the action physically plausible.

Text in the frame

Exact, legible lettering is unreliable. Paint the logo or title into the first or last frame, or add it afterwards in the video editor.

A frame in a different shape

A start image whose shape differs from the output gets stretched or cropped. Set Aspect ratio to Auto so the clip follows the image, or crop the image to the shape you need first.

Settings that matter

First frame and Last frame pin where the clip starts and ends. Duration runs from 2 to 30 seconds; find the motion at the shortest length and at 480p, then render the length and resolution you need. Generate audio adds sound in the same pass. Aspect ratio sets the shape and resolution; Auto follows the first frame. Under the advanced settings, Prompt enhancement lets the model rewrite your prompt before it runs: it helps a short prompt, but rewrites a detailed one differently on each run, so set it to Off when you have written the prompt out or need a series to match, and set a Seed to repeat a result. Above the prompt, Camera, Staging, Look and Performance take reference sets for each.

Prompts to start from

Character idle

Fixed camera. The mage breathes slowly, his cloak and hood stirring in a light breeze, the ember on his staff flickering and shedding small sparks that drift upward. Stylised painterly game art. Sound: soft wind and a quiet crackle from the ember, no background music. No dialogue, no subtitles, no text, no watermark. No character position changes, no costume changes.

Spell cast

Fixed camera, medium shot. The mage slowly raises the staff with both hands as the ember at its tip swells into a ball of fire, then thrusts it forward and the fireball shoots out of frame to the right; his cloak flaps back with the force. Stylised painterly game art. Sound: a rising whoosh and a crackling burst of flame, no background music. No dialogue, no subtitles, no text, no watermark.

Hero reveal with a push in

Slow push in toward the mage's face. The mage slowly raises the staff with both hands as the ember at its tip swells into a ball of fire, then thrusts it forward and the fireball shoots out of frame to the right; his cloak flaps back with the force. Stylised painterly game art. Sound: a rising whoosh and a crackling burst of flame, no background music. No dialogue, no subtitles, no text, no watermark.

Use it in Layer

  1. Add the approved still to a session's board.

  2. Click View on the image, then Turn into video. Create video opens in the side panel with the image as its First frame.

  3. Click the model name at the top of the panel and choose Wan 3.0. The first frame stays in place.

  4. In the prompt box, write the camera, the motion in order, the style, the sound and what must not appear.

  5. Turn on Generate audio if the clip needs sound, and leave Duration short while you find the motion.

  6. Click Create video. The cost shows on the button before you click.

  7. When the clip lands on the board, click View to open it in the viewer and play it. Then change one thing in the prompt at a time and run it again.

Learn more

Related