p-video-2-pro

Generate high-quality AI video from text or images with Pruna’s p-video-2-pro.

p-video-2-pro is Pruna’s quality-focused video model. It gives you two generation modes, control over the first and last frame, and a set of resolution and aspect ratio options so you can trade off speed against fidelity.

How it works

The model takes:

• A prompt describing the subject, action, scene, and optionally camera, lighting, and style.
• An optional image to use as the first frame for image-to-video, and an optional last_frame_image to control the ending.
• A mode, resolution, duration, and aspect_ratio.

It returns a generated video.

Inputs

• prompt (required): what to generate.
• mode: speed for faster generation (default) or quality for higher fidelity.
• resolution: 480p or 768p (default 768p).
• duration: 5 to 15 seconds (default 5).
• aspect_ratio: 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, or 1:1. Reference images determine the canvas when provided.
• image: first-frame reference image for image-to-video.
• last_frame_image: last-frame reference image for control over the ending.
• prompt_upsampler: expand your prompt before generation. off, turbo (default), or max.
• seed: set for reproducible results.

Tips

• Start with the subject, action, and scene. Add camera movement, lighting, and style when you need more precise, repeatable results.
• Use mode: speed for fast iteration, then switch to mode: quality for final renders.
• For image-to-video, use a high-quality, well-lit reference image and keep the requested motion consistent with the starting frame.
• Use last_frame_image when you need control over how the video ends.

Pricing

You pay per second of returned video, based on mode and resolution.

Resources

Docs: https://docs.pruna.ai/en/stable/docs_pruna_endpoints/index.html