How to Create AI Video from Text Prompts
Generate a video from a text description on Kensa, from writing the prompt to downloading the clip.
What you need
- A Kensa account.
- Enough credits for the model you choose (see AI Models).
Step 1: Write your prompt
Describe one shot. A good prompt covers:
- Subject: who or what is in the scene.
- Action: what happens.
- Setting: where and when.
- Camera and light: the camera move, the lighting and the style.
A golden retriever running through a sunlit meadow, cinematic slow motion, warm lighting.
Step 2: Choose a model
Open Text to Video and pick a model:
| Model | Good for | Duration |
|---|---|---|
| Veo 3.1 | Quick tests at a fixed price, audio included | 8 s |
| Seedance 1.5 Pro | 1080p and flexible length | 4-12 s |
| Grok Image Video | Cheap variations | 6 or 10 s |
| Seedance 2.0 | Longer clips with sound, on the Seedance 2.0 tool | 4-15 s |
Step 3: Set duration, aspect ratio and resolution
Use 16:9 for landscape video (YouTube, websites) and 9:16 for vertical video (TikTok, Reels, Shorts). While you test a prompt, pick the shortest duration and the lowest resolution; they cost the fewest credits. The exact cost is shown before you generate.
Step 4: Generate and download
Click Generate. Most clips are ready in a few minutes. Preview the result, download it, or find it later in My Creations. If a generation fails, the reserved credits are returned.
Tips
- Change one thing at a time. If the result is off, adjust one part of the prompt and generate again.
- Start short. Test the prompt on a short, low-resolution clip before you make the final version.
- Browse templates. Templates show prompts that work.
For a longer walkthrough, see the guide How to Create AI Video from Text Prompts.