Which AI model should I use for Text to Video?

Which AI model should I use for Text to Video?

When you use Text to Video in Powtoon Imagine, you choose between two AI (artificial intelligence) models: Seedance Pro and Veo 3. Picking the right model upfront saves you generation time and AI Credits. This article explains the difference and walks you through selecting a model for your project.

Before you start

• Open Powtoon and navigate to the Text to Video workflow in Powtoon Imagine.
• Decide whether your project needs strong visuals, generated audio, or both — this determines which model to use.
• Check your available AI Credits before generating. Each generation costs AI Credits, and the amount may vary by model.

Understanding the two models

Choosing the right AI model for Text to Video depends on what your Powtoon needs most.

Seedance Pro is optimized for visual quality. Use it when your priority is realistic motion, detailed scenes, cinematic camera direction, or high-fidelity visuals. It does not generate audio as part of the output.

Veo 3 adds native audio generation to the video output. Use it when your prompt requires synchronized sound — including spoken dialogue, sound effects, or ambient noise — as part of the generated result. Veo 3 also produces strong visuals, but its key differentiator is audio.

How to choose and use a model

1. Review your project goal. Decide whether your Powtoon is visuals-first or needs audio baked into the generated clip.
   • ► Choose Seedance Pro if you need visual realism, motion quality, or cinematic camera control.
   • ► Choose Veo 3 if your prompt includes spoken dialogue, sound effects, or soundscapes that must be part of the output.
2. Select the model in the Text to Video workflow. The model selector appears before you write or submit your prompt.
3. Write your prompt with the selected model in mind.
   • ► For Seedance Pro: focus your prompt on visual detail — describe scenes, movement, lighting, and camera angles clearly.
   • ► For Veo 3: include audio cues in your prompt — describe dialogue, tone of voice, background sounds, or specific sound effects you want generated.
4. Generate the video and review the result. Check that the visuals, motion, and (if applicable) audio match your goal.
5. If the result does not match your goal, revise your prompt and generate again — or switch models if your needs have changed.

Note: Each generation uses AI Credits. If you switch models and regenerate, additional AI Credits will be deducted. Check your AI Credits balance before generating again.

What happens next

Powtoon generates the video based on the model you selected. Seedance Pro returns a visuals-focused result. Veo 3 returns a video with synchronized audio included. Once generated, you can review the output, adjust your prompt, or switch models and try again until the result matches your project goal.