Turn a written idea or source image into a short video. Describe the subject and motion, choose the settings available for your project, and generate directly in Lanta AI.
Text and image workflows
Begin with a written scene when you need to build an idea from scratch, or use an existing image when the subject and composition are already defined. The generator keeps both paths in one workspace.
Start with the subject, action, setting, and camera move you want to see. Use the generated clip as a story opening, social hook, or visual direction, and review the settings and credit estimate shown in the form before generating.
Create From TextUpload an image that already defines the subject and composition, then describe the motion that should happen next. Try a slow camera push, moving light, or subtle background movement for a portrait, illustration, or product reveal.
Animate an ImageFor a portrait, product photo, or illustration, decide what must stay recognizable and where movement can happen before you upload it.
A person walking to the right needs space on that side of the frame. For a tightly cropped portrait or product, start with a small movement rather than a wide camera sweep.
Use a frame where the main subject is easy to separate from the background. If several objects compete for attention, crop the image around the one the video should follow.
Leave clear space for a headline or price. Add exact lettering in your editor after generation, especially when a moving label or a changing camera angle could make it hard to read.
Creative use cases
Use short generated clips to test a social opening, add motion to product imagery, or preview a story beat before committing to a longer edit or production.
Start with a short scene description, generate an opening shot for a Reel, Short, or TikTok post, then add captions, voice-over, and your final edit.
Use the image-to-video workflow to explore camera movement, lighting, or a seasonal campaign direction while keeping the product as the visual anchor.
Turn a written beat or illustration into a short scene preview, then use the result to discuss atmosphere, motion, and pacing before a larger production.
A simple workflow
Choose the right starting point, direct one clear action and camera move, then review the result and refine only what needs to change.
Choose Text to Video when your scene starts as a written idea. Choose Image to Video when a photo or illustration already defines the subject, composition, or visual direction.
Describe the main action and camera movement, then review the duration, ratio, resolution, audio, reference, access, and credit options shown in the form.
Check motion, faces, hands, text, product edges, and sound in the result. If something drifts, simplify one action or camera move before changing several settings at once.
Use these examples as starting briefs. Match the description to your own source material, then keep the first attempt focused on one visible action.
Starting point
Image to Video. Use a front-facing photo of a skincare bottle on a simple background.
Prompt
Keep the bottle upright and still. A soft reflection travels slowly across the glass while the camera moves slightly closer. Keep the cap, label placement, and background consistent. End with the bottle centered and clear space above it.
What to review
Check the silhouette and label placement before adding a headline. If the bottle warps, try a locked camera with only the lighting change.
Starting point
Image to Video. Choose a clear portrait with the eyes visible and some space around the head.
Prompt
The person blinks once, gives a small smile, and turns slightly toward the window. A few strands of hair move gently. Keep the camera still and preserve the original framing, clothing, and background.
What to review
Compare the face at the beginning and end. If a turn changes the identity, reduce it and keep only the blink and expression.
Starting point
Text to Video. Start with a setting and one event rather than a full plot.
Prompt
A small bookshop on a rainy evening. From inside, the camera faces the entrance as a visitor closes a wet umbrella and steps through the door. Warm shelves in the background, reflections on the pavement outside, a slow forward camera move.
What to review
Make sure the entrance is readable before adding another event. This shot can introduce a story; the next action can be a separate clip in your edit.
Keep the input, selected model, and settings steady while you revise one part of the prompt. That makes it easier to see which change helped.
Planning a connected sequence from one starting image? See the Wan 2.6 multi-shot workflow
FAQ
Find quick answers about starting a project, choosing an input, checking available settings, and improving the generated result.
You can turn a written scene into a short video or animate an existing photo, illustration, or product image. These workflows work well for social hooks, visual concepts, product reveals, and story previews.
Use the generator on this page, choose Text to Video or Image to Video, add your prompt or source image, review the available settings, and generate. Your completed video will appear in the generation workflow for review and download.
New Lanta AI accounts receive trial credits, but video generation is not unlimited. The form shows the current access condition and credit estimate before you generate, so review those details for the input and settings you choose.
Yes. Choose the Text to Video tab to start from a written scene or the Image to Video tab to animate an existing visual. Use the duration, resolution, ratio, audio, and reference options shown for the selected workflow.
Start with a clear image and keep the main subject easy to see. Ask for one compatible action and a restrained camera move. Review faces, hands, logos, labels, and edges, then simplify the motion if those details change.
Commercial use depends on the service, model, plan, and source material involved. Check the current Lanta AI plan terms and confirm your rights to every uploaded image, person, logo, character, and audio asset before publishing or delivering the result.
Choose Text to Video when you need to create a scene from a written idea. Choose Image to Video when a photo, product image, portrait, or illustration should define the subject and composition of the generated video.
You can adapt the same natural-language scene idea, but the result may change because each model interprets prompts differently. Preserve the subject, main action, environment, and camera direction first, then adjust one instruction or setting at a time.
The available duration and resolution options are shown directly in the generator and can vary by input and other settings. Use the current form as the source of truth before submitting a generation.
Start with a prompt or source image, focus the motion on one clear idea, and generate a clip you can review, refine, and use in a larger edit.
Start Creating