<aside> 📌 Prompts are widely used in text-to-video, image-to-video, and character-to-video generation with PixVerse. Writing accurate prompts helps PixVerse generate the desired video more efficiently, reducing trial and error. This guide explains the basic rules for prompts in PixVerse, helping you get started with AI video generation.

</aside>

1. What is a Prompt

2. Prompt Language

It is recommended to use English for input as PixVerse understands English best and yields the best results.

If you are not familiar with English, you can use your native language. PixVerse's model can handle input in any language.

3. Prompt Content

3.1 Descriptive Text

Provide detailed text describing the content you wish to generate. Use natural language and be as simple and detailed as possible to avoid ambiguity, classical language, poetry, or idiomatic expressions.

If unsure how to describe, use this formula: subject + subject description + action + environment, i.e., "What + where + doing what."

Example:

pixverse2Fmp42Fmedia2F7cb58e37-4085-4626-85b1-750dc5e95cd4_seed1321388217-ezgif.com-video-to-gif-converter.gif

3.2 Instruction Parameters

Typically include video length, resolution, frame rate, etc.

PixVerse generates videos with a default length of 4 seconds, resolution of 1408x768 (longer videos will be supported in future updates).

Upscaling can increase the resolution to 4K, significantly enhancing video quality but also increasing generation time.

                                         Upscale improves video quality

                                     Upscale improves video quality

3.3 Emotional Atmosphere

Describe the emotional tone or atmosphere of the video, useful for controlling facial expressions, environment mood, etc.

Example:

pixverse2Fmp42Fmedia2Ff418e316-a1e4-413b-934f-bebee17bbc4e_seed384091164-ezgif.com-video-to-gif-converter.gif