Vidu Q4 Image to Video: Animate Stills in 4K With Sound

Last verified: October 29, 2026

Vidu Q4 turns one still image into a video clip at up to 4K resolution. With the Generate Audio switch on, it also creates the soundtrack along with the picture. In Vidu Q4 image-to-video, the uploaded picture becomes the first frame. Your prompt then directs what happens next: the action, the camera and, when audio is on, the sound that goes with them.

Shengshu Technology built Vidu Q4 for lifelike performances. It aims for better realism and lighting with natural texture and depth, and for steady camera moves and cuts that follow what the creator asks for. Those qualities decide whether an animated still looks like real footage or like a photo that just drifts, so the prompt matters as much as the picture.

Each run here comes down to five choices: the still in Image, the wording in Prompt, the clip length in Video Duration, the output size in Resolution, and whether Generate Audio stays on. The sections below show what the model does with those choices and how to get a clean take.

Explore more image to video

Image To Video Use in Reference To Video Ltx 2 Fast Kling O3 Gemini Omni Kling 3.0 Vidu Q1 Kling 2.5 Turbo
Capability Snapshot

Inputs, Length, Resolution and Sound at a Glance

What the form accepts and what each run gives back.

Input

Image (JPG, JPEG, PNG or WEBP, up to 10 MB) + a text prompt of up to 5,000 characters

Video Duration

3 to 16 seconds

Resolution

540P, 720P, 1080P, 2K or 4K

Audio

Switch on or off, on by default

Output

1 video per run

How do you write a Vidu Q4 prompt that moves well?

Prompt the Motion, Not the Picture

Vidu Q4 uses the uploaded image as the first frame of the clip, so the look of the shot is already set before your prompt is read. Use the prompt for what changes: who moves, how they move and what the camera does. Vidu describes this mode the same way, as one image plus a description of the motion and the camera. Re-describing the subject's clothes or the background takes up characters without telling the model anything new.

Give Every Movement a Direction and a Speed

Vidu's motion-prompting advice is that instructions along one axis, with a direction and a speed word, give smoother motion than broad phrases like dynamic movement through the scene. Write a slow push to the left, a gentle upward tilt or a slight head turn to the right, rather than leaving the model to invent the action. Add a word like subtle, gentle, slow or steady; Vidu notes these words make the motion vary less from take to take.

Start From a Sharp, Well-Lit Image

For image-to-video, Vidu recommends high-quality source images with clear subjects and good lighting. Before you upload, pick the version of the still where the subject is in focus and stands out from the background. Treat the upload's minimum size as a floor, not a target: if the clip is meant for a high resolution, start from a photo that already holds that much detail.

How do you animate a photo with Vidu Q4?

Five steps take you from a still to a downloaded clip.

1

Add your still in "Image"

Drop the picture into "Image". The run will not start without it. Choose the exact frame you want the video to open on, not a close variant, because the clip starts from that image. If the upload is refused, check the file against the format, size and minimum-dimension rules in the FAQ.

2

Describe the motion in "Prompt"

The "Prompt" box is required and takes up to 5,000 characters. Write what should happen after the opening frame: the action, the camera move and, if sound is on, what the viewer should hear. If all you have is a rough idea, click "Enhance Prompt" under the box for suggested wording.

3

Set "Video Duration" and "Resolution"

Pick a length in "Video Duration", from 3 to 16 seconds (5 by default), and an output size in "Resolution": 540P, 720P, 1080P, 2K or 4K, with 720P selected to start. Test the motion with short clips at a lower tier, then switch to 2K or 4K for the take you will deliver. That way the high-resolution render goes to a prompt you have already tested. Also check the "Generate Audio" switch. It is on by default, so leave it on if the clip should come back with sound.

4

Start the run with "Generate"

Click "Generate" when the settings look right. Each run returns one video. To compare two ideas, change one thing at a time (the prompt, the length or the resolution) and run again, so you can tell which change made the difference.

5

Review and download the result

Watch the clip with sound before you save it. Check that the motion ends where the prompt said it would, and that any dialogue or effects fit the scene. If it holds up, download it. If not, adjust the prompt and generate again.

Fixing Common Problems With Animated Stills

Problems that come up when animating a still, and what to change before the next run.

The clip ends before the action I described finishes.

Video Duration sets the total length of the clip and starts at a short default, so a prompt with several beats can run out of time. Raise Video Duration toward its longest option, or cut the prompt down to the one action the clip has to show. No single run goes past the longest option. For a longer sequence, export a frame from the end of the first clip, upload it in Image for the next run, and join the clips in an editor. Generate again after changing the length or the prompt.

There's talking or sound effects in my clip that I didn't ask for.

With Generate Audio on, Vidu Q4 returns video with sound, and that sound can include dialogue and sound effects. If you plan to add music or a voiceover later, switch Generate Audio off and run again: the model then outputs a silent clip. If you want sound but a different kind, spell it out in the prompt (the ambience, the effects and whether anyone speaks) and generate again.

The face drifts, or the same glitch shows up on every take.

Vidu's guidance separates flaws worth simply re-running from flaws that need new inputs. If the problem is small and in one place, like flicker in a corner or a single bad frame, and the motion otherwise looks right, generate again with the same image and prompt. If the motion itself is wrong, the face drifts, or the same artifact appears in the same spot across two or more takes, change the inputs before the next run: simplify the movement in the prompt or upload a clearer image.

The motion jitters when the subject and the camera both move.

Opposite directions are a common cause. Vidu notes that a character walking right while the camera tilts left often comes out as jitter instead of a deliberate shot. Rewrite the prompt so the camera and the subject move in compatible directions, or keep one of them still, then generate again.

The prompt box won't take any more text.

The Prompt box has a character limit, and long prompts usually hit it by describing what the image already shows. Cut the description of the subject and the setting, since the still already covers both. Keep the action, the camera and the sound, then generate.

Frequently Asked Questions

Do I need an image, or can Vidu Q4 start from a text prompt alone?

You need both. Image and Prompt are required, and every run starts from the picture you upload. If all you have is a written idea, make or choose a still first, then bring it here and use the prompt to describe how it should move.

Which image files can I upload, and how large can they be?

The Image upload accepts JPG, JPEG, PNG and WEBP files up to 10 MB, and the picture must be at least 300×300 pixels. That minimum is only the entry point. If you plan to render at one of the higher Resolution options, start from a photo with far more detail than that.

Does Enhance Prompt rewrite my prompt automatically?

No. Enhance Prompt, the button under the prompt box, suggests fuller prompts based on a rough idea, and your text only changes when you pick one of those suggestions. It's a quick way to get a first draft of action and camera wording, which you can then edit to match your still.

What makes one generation cost more than another?

Once you are signed in, the credit cost of a run is calculated from the options you picked and shown above the Generate button. Two settings change it: Video Duration and Resolution. Picking a longer clip or a higher resolution changes the figure you see. Using Enhance Prompt does not use credits.

References

Sources and citations used to support the content provided above.

Updated: 2026-10-29 21:17:49 4 Sources

platform.vidu.com

Source Link
https://platform.vidu.com/docs/api-reference/video-models/vidu-q4-preview/image-to-video

www.vidu.com

Source Link
https://www.vidu.com/vidu-q4

www.vidu.com

Source Link
https://www.vidu.com/ai-image-to-video

www.vidu.com

Source Link
https://www.vidu.com/blog/motion-control-ai