Guide›Media Studio
Media StudioBeginner5 min read

Generating video

Generate video from text or from an image through fal.ai models, tune length, motion, and quality, then preview the result and export it to your asset library or a project.

Two ways in

Myndlab Media Studio supports two ways to generate video through fal.ai models: text-to-video, where you describe the scene in words; and image-to-video, where you start from a still image and add motion to it. Choose the entry point based on what you already have.

💡
Tip.If you have an image in the look you want, start with text-to-image in the studio, then pass the result into image-to-video — that gives you tighter control over the look.

Text-to-video

In text-to-video mode, you describe both the scene and the motion. Name the subject, what it does, and how the camera moves:

text
A paper boat drifting down a rain-soaked street gutter at night,
neon reflections on the water, slow dolly-in, gentle ripples.

Motion verbs (drifting, floating, panning) and camera words (dolly, orbit, static) are what separate a video prompt from an image prompt. Be explicit about speed and direction.

Image-to-video

In image-to-video mode, you start from a still frame — an image you generated in the studio or uploaded — and describe only the motion you want added to it:

1
Pick the source image
Use an image from the asset library or upload one. This image defines the first frame and the whole look of the clip.
2
Describe motion only
Since the look is already set by the image, your prompt is just the motion: a light breeze, a forward camera push, a subtle ripple.
3
Keep the motion modest
Subtle, consistent motion looks more convincing than large motion, which can distort elements in the source image.

Length, motion & quality

Three options control the nature of the resulting clip, and together they affect how long generation takes:

1
Length
The clip duration. Shorter clips generate faster and tend to stay more consistent; start short to confirm the direction before going longer.
2
Motion amount
How strong the motion in the scene is. Higher motion is livelier but more prone to artifacts; lower motion is calmer and more stable.
3
Quality
Higher-quality settings produce cleaner detail but take longer. Use lower quality for drafts and the highest setting for the final pass.
ℹ
Note.Video generation is much slower than image generation. Preview at low settings and short lengths while experimenting, and reserve high settings for the final take only.

Previewing & exporting

Once generation finishes, you can play the clip inside the studio to confirm the motion and timing match your intent before exporting.

1
Preview the clip
Play the result and check how the motion begins and ends. If it is off, adjust the prompt or lower the motion amount and regenerate.
2
Export as MP4
Download the clip as MP4 to use anywhere, or ship it straight into a Web App or Design project as an asset.

All generated clips stay in the asset library, so you can return to any earlier run or compare several attempts before choosing the final version.