Official BytePlus partnerBytePlus, a ByteDance company
Text to video · Official BytePlus partner
Text to video AI: type a scene, get a video
Describe a shot in plain words and Seedance renders it with the camera move, light and motion you asked for. From $0.063 per second, 720p and 1080p, no watermark.
- Seedance 2.5
- Seedance 2.0
- Seedance 2.0 Fast
- Wan 3.0
- 720p and 1080p
- 4 to 15 s
Last updated
On this page
Short answer
Text to video AI turns a written description into a video clip. On KreatorFlow you type one scene (subject, action, place, light and camera move), pick a model and get a 4 to 15 second clip with no watermark. Use Seedance 2.5 for final shots, Seedance 2.0 for everyday clips and Seedance 2.0 Fast for cheap drafts; Wan 3.0 is the lowest price per second. Prices start at $0.063 per second at 720p.
Which model to use for a text-only prompt
Price per second for a 5-second clip, from the credits the studio charges, on the Max plan (the lowest regular price per credit).
| Model | 720p per second | 1080p per second | Credits, 5 s clip | Use it for |
|---|---|---|---|---|
| Seedance 2.5 | $0.190 | $0.396 | 24 (720p) / 50 (1080p) | Final shots from words alone: the newest Seedance model, with clips up to 30 seconds. |
| Seedance 2.0 | $0.119 | $0.238 | 15 (720p) / 30 (1080p) | Everyday text to video: cinematic camera moves and lighting from a one-paragraph prompt, at a lower price than 2.5. |
| Seedance 2.0 Fast | $0.095 | $0.158 | 12 (720p) / 20 (1080p) | Drafts: test a prompt quickly and cheaply, then render the keeper on 2.0 or 2.5. |
| Wan 3.0 | $0.063 | $0.127 | 8 (720p) / 16 (1080p) | Simple scenes, landscapes and stylised looks at a low price per second. |
Every plan renders 720p and 1080p, and the studio shows the credit cost before you generate. Plans with a higher price per credit cost more per second; see the AI video cost calculator for your plan and clip length.
The text to video prompt formula
Write one scene, in this order. One subject, one action and one camera move per clip.
- 1
Subject
Who or what is on screen, with one or two concrete details: "a small red fox in a green knitted scarf", not "an animal".
- 2
Action
One continuous action that fits 4 to 15 seconds: "trots through the snow, snow puffing up under its paws".
- 3
Place and light
Where and when, and the light: "a snowy pine forest at sunrise, soft golden backlight". Light changes the mood fastest.
- 4
Camera
One camera instruction in film terms: low tracking shot, slow push-in, crane up, orbit, handheld. Say how fast and where it ends.
- 5
Look
Finish with the texture of the picture: "shallow depth of field, 35mm film look" or "moody cinematic colour grade".
Pick the aspect ratio in the settings, not in the prompt (16:9, 4:3, 1:1, 3:4, 9:16, 21:9). Test the prompt at 720p or on Seedance 2.0 Fast, then render the keeper at 1080p. For more prompts by shot type, open the Seedance prompt builder.
Three text to video examples
Each clip was made on KreatorFlow from the prompt under it alone: no image or video reference, 16:9, rendered at 1080p, 5 seconds, sound off.
Fox in the snow · Seedance 2.5
“A small red fox wearing a knitted green scarf trots through a snowy pine forest at sunrise, snow puffing up under its paws. Low tracking shot at the fox's eye level moving alongside it, soft golden backlight through the trees, drifting snowflakes, shallow depth of field, cinematic nature film look.”
Coffee by a rainy window · Seedance 2.0
“A plain white ceramic coffee cup on a worn wooden cafe table, steam curling up from the coffee, rain running down the window behind it. Slow push-in from a medium shot to a close-up of the steam, warm morning window light, shallow depth of field, 35mm film look.”
Paper boat in the rain · Seedance 2.0 Fast
“A folded paper boat floats down a rain-filled street gutter at dusk, drifting past glowing orange reflections of street lamps in the puddles. Low camera at water level tracking alongside the boat, raindrops rippling the water, moody cinematic colour grade.”
Common text to video mistakes
- Several scenes in one prompt: the model makes one continuous shot. Make one clip per scene and join them
- Feelings instead of pictures: "an emotional moment" shows nothing; "her eyes fill with tears as she reads the letter" does
- No camera direction: without one the model picks the movement for you
- Long, repeated text: one extra visual detail helps more than saying the same thing twice
Have a picture already?
If you already have the image, start from it in image to video AI: the picture fixes the subject and the look, and the prompt only has to describe the motion. For a character who appears in many clips, build them once with the AI character generator.
KreatorFlow is an official partner of BytePlus, a ByteDance company. Powered by ByteDance's Seedance 2.0 and Seedream models via BytePlus, a ByteDance company.
Text to video questions
What is text to video AI?
A model that makes a video from a written description alone, with no photo or clip to start from. You describe what is on screen, what happens and how the camera moves, and the model renders it as a short clip.
Which model should I use for text to video?
Seedance 2.5 for final shots that need the most detail and longer clips. Seedance 2.0 for most everyday clips. Seedance 2.0 Fast to test a prompt cheaply before you render the final version, and Wan 3.0 when price per second matters most on simple scenes and stylised looks.
How much does text to video cost per second?
On the Max plan, which has the lowest regular price per credit, Seedance 2.0 costs $0.119 per second at 720p and $0.238 at 1080p; Seedance 2.5 costs $0.190 and $0.396; Seedance 2.0 Fast $0.095 and $0.158; Wan 3.0 $0.063 and $0.127. The studio shows the exact credits before you generate, and failed generations are refunded.
Can I try text to video for free?
New accounts get 25 starter credits for images and video. Free accounts can make video on Seedance 2 Fast and Seedance 2.0 Mini at 720p, queued and rate limited. A plan adds every video model, 1080p and above, and no queue.
How long can a text to video clip be?
Seedance 2.0 makes 4 to 15 second clips and Seedance 2.5 goes up to 30 seconds. One prompt gives one continuous shot; for a longer story, make several clips and join them in the KreatorFlow editor.
Can text to video clips have sound?
Yes. Seedance can generate ambient sound and effects that match the scene. Turn audio on in the studio, or leave it off for a silent clip like the examples on this page.
Is there a watermark?
No. Clips are rendered with the watermark off and KreatorFlow adds no logo of its own.
Turn your first prompt into a video
Open the video studio, type a scene and generate. New accounts get 25 starter credits for images and video.