AI-powered resolution enhancement for images or video to restore clarity.
# Image Upscale (P) High-speed AI resolution enhancement for professional imagery. ## Overview Image Upscale (P) is a high-performance resolution enhancement tool designed to bridge the gap between low-resolution source files and high-fidelity marketing assets. By leveraging advanced neural networks, this model intelligently reconstructs pixel data, effectively increasing the dimensions of images while maintaining structural integrity and sharpness. Whether you are working with legacy assets, web-scraped thumbnails, or compressed social media exports, this tool provides a rapid solution to restore clarity and professional-grade quality. What distinguishes Image Upscale (P) from traditional interpolation methods is its speed and precision. Capable of processing images in under one second, it is the fastest upscaler currently available for production-grade workflows. It supports output resolutions up to 8 megapixels, ensuring that your final assets are print-ready or suitable for high-density digital displays. The model offers granular control over the upscaling process, allowing users to toggle between factor-based scaling and target-megapixel resolution, making it highly adaptable to specific project requirements. ## Use Cases * **E-commerce Product Optimization:** Upscale low-resolution manufacturer images to meet high-definition standards for product detail pages. * **Social Media Asset Restoration:** Enhance blurry or pixelated user-generated content for use in high-impact ad campaigns. * **Print-Ready Conversion:** Increase the resolution of digital graphics to ensure they meet the DPI requirements for physical marketing collateral like brochures and posters. * **AI-Generated Art Refinement:** Improve the texture and clarity of AI-generated images that often suffer from soft edges or low-resolution artifacts. * **Legacy Archive Digitization:** Restore clarity to older, lower-resolution brand assets to maintain visual consistency across modern marketing channels. * **Billboard and Large Format Display:** Prepare digital assets for large-scale displays by maximizing pixel density without sacrificing visual quality. ## Parameters This model offers a robust set of configuration options to fine-tune your output. Please refer to the input parameters section in the Pixloop interface to adjust settings such as scaling factors, target megapixel resolutions, output formats, and enhancement toggles for realism and detail. ## Tips for Best Results * **Use 'Enhance Realism' for AI Content:** If you are upscaling images generated by models like Midjourney or DALL-E, enable the 'enhance_realism' flag to ensure textures remain natural and avoid the 'plastic' look often associated with standard upscaling. * **Prioritize PNG for Quality:** When output quality is paramount, select 'png' as your output format to avoid the compression artifacts associated with lossy formats like JPG. * **Balance Detail vs. Accuracy:** Use 'enhance_details' sparingly; while it sharpens textures, it can sometimes introduce minor deviations from the original source material. Use it primarily for images where texture definition is more important than pixel-perfect fidelity. * **Target Resolution vs. Factor:** Use the 'target' mode when you have a specific pixel count requirement (e.g., 4MP for standard web banners), and use the 'factor' mode when you simply need to double or triple the current size of an asset. * **Monitor Output Quality:** If you are using JPG or WebP formats, ensure your 'output_quality' is set to 90 or above to minimize compression noise during the final save process. ## About Image Upscale (P) is provided by PrunaAI and is hosted via Replicate. It represents a significant advancement in real-time image processing, utilizing optimized neural architectures to deliver industry-leading speed without compromising on output quality. Designed for high-throughput environments, this model is a staple for creative teams requiring rapid, reliable, and high-quality image enhancement.
Model input reference derived from preset schema.
| Name | Type | Required | Default | Description |
|---|---|---|---|---|
| image | image | Yes | null | Input image to upscale. |
| no_op | boolean | No | false | Health check mode - returns status without inference. |
| factor | number | No | 2 | Scaling factor applied to each side of the image (used when upscale_mode is 'factor'). Output capped at 8 MP. |
| target | integer | No | 4 | Target resolution in megapixels (used when upscale_mode is 'target'). |
| upscale_mode | select | No | "target" | Upscale mode. 'target' scales to a fixed megapixel resolution. 'factor' multiplies each side by the given factor. |
| output_format | select | No | "png" | Format of the output images |
| output_quality | integer | No | 90 | Quality when saving the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. Not relevant for .png outputs |
| enhance_details | boolean | No | false | Enhance fine textures and small details. May increase contrast and introduce minor deviations from the original image. |
| enhance_realism | boolean | No | false | Improve realism. May deviate more from the original image. Recommended for AI-generated images. |
| disable_safety_checker | boolean | No | true | Disable safety checker for generated images. |