Real-ESRGAN vs Ximilar Image Upscaler
Real-ESRGAN
github.comXimilar Image Upscaler
ximilar.comSide by side
| Figure | Real-ESRGAN | Ximilar Image Upscaler |
|---|---|---|
| Price from | — | $69/mo |
| Free plan | — | Yes |
| Free trial | — | — |
| Open source | Yes | — |
| Web | Yes | Yes |
| Windows | Yes | — |
| macOS | Yes | — |
| Linux | Yes | — |
| API | — | Yes |
| Self-hosted | Yes | Yes |
| Plans published | 0 | 16 |
| Spec score · Image Upscaling Software | 8.2 (#33) | 9.0 (#6) |
| Spec score · AI Image Upscalers | 7.6 (#24) | 8.4 (#13) |
| Batch processing | Yes | The service is designed to upscale image collections in bulk through its REST API. |
| Free plan | Yes | Yes |
| Models | The project provides models for general images, anime images, and animation video. | The API offers a Classic model for lower cost and a GAN model for higher-quality upscaling with improved detail. |
| Output formats | The executable can output JPG, PNG, and WebP images. | API requests can return PNG, JPG or WebP output, with automatic format preservation available. |
| Purpose | Real-ESRGAN develops practical algorithms for general image and video restoration, extending ESRGAN using training with synthetic data. | The Image Upscaler uses AI-powered super-resolution to upscale images in bulk. |
Plans
Ximilar Image Upscaler
- FreeFree
- Business 100K$69 a month
- Professional 1M$579 a month
- Professional 3MPrice not published
- Professional 4MPrice not published
- Professional 5MPrice not published
- FreeFree
- API credit pack 10,000$11, paid once
- Business 100K$69 a month
- Business 300K$199 a month
In their own words
Real-ESRGAN
Real-ESRGAN is a free, open-source project developing algorithms for image and video restoration. It offers pretrained models for general images, anime images and animation video, including AnimeVideo-v3 and smaller anime models. Image inputs include JPG, PNG and WEBP; inference can handle alpha-channel, grayscale and 16-bit images, and outputs can be JPG, PNG or WebP. Portable inference supports scale ratios of 2, 3 or 4, with 4 as the default; the Python implementation also offers arbitrary output scaling. Portable NCNN executables are available for Windows, Linux and macOS, with required binaries and models included and no CUDA or PyTorch environment needed. These executables do not include every Python feature, including outscale. Python use requires Python 3.7 or newer and PyTorch 1.7 or newer. Online inference is available through an ARC demo and Colab demos. GFPGAN is integrated for face enhancement, and the training code supports fine-tuning on user data or paired data. The project uses synthetic training data and is released under the BSD-3-Clause license.
Ximilar Image Upscaler
Ximilar Image Upscaler has no write-up on Specifiction yet; its figures are above.