@mashy Thank you for the kind words! We are thrilled to introduce our Adima AI image upscaler, which indeed has the capability to upscale images up to 16x while maintaining excellent quality. Our software achieves this through the use of Generative Adversarial Networks (GANs), a cutting-edge AI technology known for its ability to generate high-quality, realistic images. Here’s a bit more detail on how we maintain image quality during such significant enlargement: How We Maintain Image Quality at High Levels of Enlargement : GAN Technology: Generative Adversarial Networks (GANs) are the core of our upscaling process. They work by pitting two neural networks against each other: a generator, which tries to create high-resolution images, and a discriminator, which attempts to distinguish between real high-resolution images and the generated ones. This adversarial process helps the generator improve over time, producing increasingly realistic and detailed images. Tile-Based Processing: Our upscaler enhances image quality by breaking down images into smaller, manageable tiles. Each tile is processed individually, allowing for finer detail and better control over the upscaling process. This approach ensures that even at 16x enlargement, the image retains its sharpness and clarity across all regions, avoiding the common pitfalls of blurriness and artifacting that often accompany such high levels of scaling. Preservation of Details: By focusing on smaller sections of the image, our model can pay attention to the intricate details and textures in each tile. This meticulous approach allows the upscaler to maintain the original character and quality of the image, making it particularly effective for detailed images such as photographs, artworks, and technical diagrams. Future Plans for Expansion : We are indeed planning to expand the compatibility of our software to other operating systems. Currently, our focus is on the successful launch of the software for its primary platform, but: Linux and macOS Support: We are committed to expanding our software’s compatibility to both Linux and macOS. The feedback from our initial launch will guide our development process, ensuring that we deliver a robust and user-friendly experience on these platforms. This expansion is a high priority for us following the successful rollout on our current platform. Ongoing Development and Future Enhancements We are also working on training a new model with a significantly larger dataset, which is expected to be a game-changer in the field of image upscaling: Training on a Large Dataset: Our new model is being trained on a substantial dataset comprising between 100,000 and 500,000 images. This extensive dataset will enable the model to learn from a diverse range of image types and qualities, enhancing its ability to upscale images even more effectively and with greater fidelity to the original content. Anticipated Improvements: The enhanced model will likely offer superior performance in terms of upscaling quality, speed, and versatility across different types of images. We anticipate that this will set a new benchmark in the industry for image upscaling solutions. Thank you for your interest, and we look forward to sharing more updates as we continue to develop and expand our software. Your support and feedback are invaluable as we strive to bring innovative solutions to the world of digital imaging.