Bigjpg vs Real-ESRGAN
Bigjpg
bigjpg.comReal-ESRGAN
github.comSide by side
| Figure | Bigjpg | Real-ESRGAN |
|---|---|---|
| Price from | Free | — |
| Free plan | Yes | — |
| Free trial | — | — |
| Open source | — | Yes |
| Web | Yes | Yes |
| Windows | Yes | Yes |
| macOS | Yes | Yes |
| Linux | — | Yes |
| iOS | Yes | — |
| Android | Yes | — |
| API | Yes | — |
| Self-hosted | — | Yes |
| Plans published | 4 | 0 |
| Spec score · Image Upscaling Software | 9.1 (#2) | 8.2 (#33) |
| Spec score · AI Image Upscalers | 9.1 (#2) | 7.6 (#24) |
| Batch processing | Yes | Yes |
| Desktop app | Yes | Yes |
| Free plan | Yes | Yes |
| Maximum scale | custom | custom |
| Purpose | Bigjpg uses deep convolutional neural networks for single-image super-resolution to enhance user-provided images. | Real-ESRGAN develops practical algorithms for general image and video restoration, extending ESRGAN using training with synthetic data. |
| Support | Bigjpg lists [email protected] for feedback, contact, and support. | The README directs questions to the listed email addresses [email protected] and [email protected]. |
| Upscaling method | ai | ai |
Plans
In their own words
Bigjpg
Bigjpg enlarges user-provided images using deep convolutional neural networks. It can scale images up to 16x, reduce noise and compression artifacts, and enhance detail and sharpness. Users can choose photo or artwork styles, set enlargement from 2x to 16x, and select noise reduction from None to Highest. Batch processing and API access are listed, and the service links to Windows, Mac, iOS, and Android applications. The free plan allows 20 pictures per month, with a 5MB maximum upload and enlargement up to 4x. Paid plans allow uploads up to 50MB and enlargement up to 16x: Basic is $6 a month, billed 2 Months as a one-time payment; Standard is $12 a month, billed 6 Months as a one-time payment; Premium is $22 a month, billed 12 Months as a one-time payment. Bigjpg says uploaded images are used to provide the requested enhancement and deleted automatically within 24 hours for free accounts or 72 hours for premium accounts. It says uploads are not used for model training without explicit, informed consent. Users who are not logged in must keep the browser open during processing or the enlarged result will be lost.
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.