FlexPicOpen the app

Text to video AI

Type what you want to see and get a short video. Try the same prompt on different models and keep the best take.

A subscription is required for new accounts. Video generation needs a subscription.

What you can make

How it works

  1. Open Text to Video and write your prompt.
  2. Choose the model, duration, aspect ratio and resolution.
  3. Generate and download the video.

Models available here now: Seedance 2.0, Kling 3.0, Kling 3.0 Pro, Kling O3, Kling 2.5 Turbo, Kling 2.5, Veo 3.1 Fast, Hailuo 3, Gen-4.5 and Wan 2.7 and more. The list in the app is always the current one.

Simple credit pricing

Credit packs start at $10 (2,300 credits) and subscriptions start at $14 per month (3,100 credits). See all prices.

Write a text-to-video prompt as a single shot

A useful prompt tells the model what is in the frame and what changes over time. Begin with a shot size, identify one subject, give it one action, then describe the camera and lighting. Avoid asking a short clip to cover a chase, a conversation and a scene transition at once.

For example, replace “a beautiful mountain scene” with “wide shot of a mountain lake at dawn; mist drifts over the water as the camera slowly pans right; soft light and natural colors.” The second description makes the movement and framing explicit without turning the prompt into a long list of competing instructions.

Three prompts to adapt

  • Product: A glass perfume bottle on pale stone. A narrow band of reflected light moves across the glass while the camera slowly orbits. Clean studio background, realistic materials.
  • Environment: Wide shot of a quiet train platform at blue hour. A light breeze moves a paper sign; the camera remains still. Gentle ambient light, one continuous shot.
  • Character: Medium shot of an illustrated explorer in a bright yellow coat. The explorer turns toward the horizon and smiles; a slow camera push-in. Soft overcast light, consistent character design.

These are starting prompts, not guaranteed outputs or demonstrations of a particular model. Keep your first take simple, inspect it, and change one instruction at a time.

Refine the result without changing everything

If motion is too strong, ask for smaller movement or a static camera. If the scene changes unexpectedly, remove secondary subjects and conflicting style terms. If a product shape or character identity matters more than inventing a new scene, switch to image to video with a reference picture you have permission to use.

Check the chosen model’s duration and aspect options before rendering. The studio shows the request’s credit price so you can compare takes with the same settings. New accounts need a subscription; there are no signup credit grants.

Use the video prompt builder, read the five-part prompt guide, or return to the AI video generator overview.

Frequently asked questions

How do I write a good video prompt?

Describe the subject, the action, the camera movement and the lighting in one or two sentences. Start simple, then add detail based on the first result.

Which text to video models are included?

Currently: Seedance 2.0, Kling 3.0, Kling 3.0 Pro, Kling O3, Kling 2.5 Turbo, Kling 2.5, Veo 3.1 Fast and Hailuo 3 and more.

Can the video include sound?

Some models can generate audio with the video; the audio switch appears in the tool when the chosen model supports it.

Are signup credits included?

No. New accounts start with zero credits and need an active subscription to generate images, video or audio, use AI chat, or write storyboards. Existing balances are preserved. Credit packs are top-ups after subscribing.

How much does a generation cost?

Every generation uses credits. The amount depends on the model, length and resolution and is calculated by our server, not by your browser. Credit packs start at $10 for 2,300 credits, and subscriptions start at $14 per month. See the pricing page.

What happens if a generation fails?

If a generation fails on our side the credits for that job are refunded to your balance automatically.