Overview

P-Image Upscale is an AI-powered image upscaling model developed by Pruna. It enhances image resolution while preserving detail and realism.

The model lets you upscale images to a specified target resolution (1–8 megapixels), perfect for improving low-resolution assets, refining AI-generated images, and preparing visuals for production use.

How it Works

Prompt Interpretation

P-Image Upscale operates directly on input images rather than text prompts. It interprets the structure, textures, and details of the source image to reconstruct a higher-resolution version while maintaining visual fidelity.

Optional enhancement parameters can guide the model toward sharper detail or more realistic rendering.

Image Upscaling

The model increases image resolution using AI-based reconstruction. You can define a target resolution in megapixels, and the model scales the image accordingly while preserving composition and structure.

It also supports optional enhancements:

• Detail enhancement to sharpen textures and fine elements
• Realism enhancement to improve photographic quality and depth

Image Quality and Reconstruction

P-Image Upscale applies intelligent reconstruction to maintain clarity during scaling. It enhances edges, textures, and small features while minimizing artefacts typically introduced by traditional upscaling methods.

This makes it particularly effective for improving compressed assets or low-resolution inputs.

Key Features

• AI-Based Resolution Scaling

  Upscales images to a target resolution between 1 and 8 megapixels.

• Detail Enhancement

  Improves fine textures and small visual elements.

• Realism Enhancement

  Enhances lighting, depth, and photographic quality.

• Flexible Output Formats

  Supports multiple formats including JPG, PNG, and WebP.

• Production-Ready Performance

  Designed for fast, scalable workflows with high request throughput.

How to Use

1. Upload or provide the source image.
2. Specify the target resolution (in megapixels).
3. Optionally enable detail and/or realism enhancement.
4. Run the upscaling process.
5. Adjust parameters if needed to refine output quality.

For best results, use higher-quality input images and enable enhancements when working with AI-generated or compressed visuals.

Example usage:

"Upscale a 1024×1024 AI-generated image to 4 MP with enhanced details and realism, output as PNG for high-quality rendering."

Documentation

You can find full usage details, parameters, and examples here:

https://docs.api.pruna.ai/guides/models/p-image-upscale