How do I generate AI video from text (text-to-video)?

On Kunya, go to Dashboard → Videos, type your prompt, select a model like Sora 2 or Kling V3, choose duration and aspect ratio, and click Generate. Your video renders in the cloud and appears in your library when complete — no GPU or local setup needed.

Step by step

  1. Open Dashboard → Videos on Kunya
  2. Choose Text to Video mode
  3. Write a detailed prompt — describe the scene, motion, camera angle, and style
  4. Select your model:
    • Sora 2 — best for cinematic quality and prompt adherence
    • Kling V3 — fast, realistic, great for UGC-style clips
    • Kling O3 — best detail and physics for complex scenes
    • Wan 2.6 — most cost-efficient for high-volume generation
  5. Choose duration (5s, 8s, 10s) and aspect ratio (16:9, 9:16, 1:1)
  6. Click Generate and wait — most clips render in 60–120 seconds

Prompt tips for better video

  • Include camera movement: "slow dolly forward", "orbit around subject", "static wide shot"
  • Describe lighting: "golden hour", "studio lighting", "moody blue ambient"
  • Keep subjects clear: one or two main subjects render more consistently than crowds
  • Use style anchors: "cinematic 4K", "DSLR documentary", "anime style"

Related

More answers

Try Kunya free →← All answers