AI text to video generator online
Generate a short video from text
Epic fantasy shot: a colossal dragon made entirely of ocean water rises from a stormy sea beside a lone stone lighthouse. Spray and foam pour from its wings as it circles the tower. The camera orbits from sea level upward, thunder rolling, dramatic silver moonlight, highly detailed water simulation.
Try this promptWhat is Text to Video?
Describe a shot in a sentence and get a short video clip with sound. The text to video generator turns a written scene into 4 to 30 seconds of footage: the subject, what it does, how the camera moves and the mood of the light. Pick a duration, an aspect ratio and a model, hit Generate and watch the queue while the clip renders.
Two model families are available on one page. ByteDance Seedance 2.5 and 2.0 handle multi-shot storytelling, go up to 1080p and 15 seconds (30 on 2.5), and generate sound effects and voice from the prompt; the Fast and Mini versions render the same scenes cheaper. Google Gemini Omni Flash brings real-world knowledge and physics, native dialogue, clips up to 15 seconds and upscaled 1080p or 4K, at a low per-second price. Send the same prompt to each and keep the one that fits.
What you can make with Text to Video
Sound included
Dialogue, ambience and effects are generated with the picture. Write a line as says: "..." and the character speaks it.
Create a 30-second, 1080p ultra-realistic documentary-sty
4 to 30 seconds
Short loops or a full beat of story. The Seedance 2.0 family goes from 4 to 15 seconds and 2.5 up to 30; Gemini Omni Flash goes from 3 to 15.
A cinematic, high-action fantasy video sequence
Two model families, seven versions
Seedance for multi-shot scenes, 1080p and long clips, with Fast and Mini for cheap drafts; Gemini Omni Flash for believable physics, dialogue and 4K at a low per-second price.
Create a 15-second cinematic fantasy slice-of-life sequen
Every aspect ratio
16:9, 9:16, 1:1, 4:3, 3:4 and 21:9 on Seedance; 16:9 and 9:16 on Gemini Omni Flash.
0:00–0:05 — LONELY HIGHWAY Wide cinematic establishing shot of a remote two-lane highway i
Video takes longer than images. Queue position and the estimated finish time stay visible, and the clip lands in your library when done.
Write slow dolly in, handheld, orbit or drone shot and the models follow. One or two actions per clip keeps motion clean.
Made by the community
Seedance videos shared by creators. Hover to play, open one to see the full prompt and recreate it.
From idea to video in three steps
- 01Describe the shot
Write the subject, the action and the camera move, then the setting and mood. Add dialogue as says: "..." if you want speech. Keep it to one or two actions per clip.
- 02Set duration, ratio and model
Choose how long the clip runs and whether it is vertical, horizontal or square. Pick Seedance Mini or Gemini Omni Flash at 360p for a cheap draft, Seedance 2.5 for longer scenes, Gemini Omni Flash for dialogue and physics. The cost shows on the button.
- 03Generate and download
Hit Generate and follow the queue. When the clip finishes it plays in the panel on the right; download the MP4, or re-run with a tweak to the prompt.
- Order the sentence as subject, action, camera, setting, mood. Models read the first words most carefully.
- Ask for one or two actions per clip. A 5-second shot with five beats comes out rushed and blurry.
- Name the camera move: slow dolly in, handheld, orbit, static wide. Vague words like "dynamic" give unpredictable motion.
- For speech, write says: "the exact line". Short lines fit the clip; long speeches get cut off. English prompts give the most reliable results.
Try these prompts
Click one to drop it into the generator above, then tweak a few words.
A matte white sneaker on a rotating pedestal in a dark studio, slow 360-degree orbit, soft rim light tracing the edges, dust particles in the air, premium commercial feel
Try itNature drone shotDrone shot gliding over a misty pine forest at sunrise, slow forward push, golden light breaking through the fog, a river winking below, calm cinematic mood
Try itCharacter with dialogueA barista in a small cafe looks up from the espresso machine, smiles and says: "Your usual?", handheld medium shot, warm morning light, shallow depth of field
Try itUrban night sceneA woman in a yellow raincoat walks through a neon-lit alley in the rain, camera tracking from behind at shoulder height, reflections on wet pavement, moody synthwave atmosphere
Try itBuilt for these creators
Vertical 9:16 clips for Reels, TikTok and Shorts, with sound already attached. Draft on Seedance Mini, finish on Seedance 2.5 or Gemini Omni Flash.
Product reveals, loops for listing pages and short ad cuts without a shoot. Pair with image to video to animate your own product photos.
Storyboard-quality previews of a concept before booking a crew. Test three moods for the same script in an afternoon.
Short visual intros, scene setters and B-roll for lessons and presentations, generated from the paragraph you are already writing.
Text to Video FAQ

Try Text to Video now
No prompt skills needed. Write a sentence, pick a model and generate; the first images are on us.

