realesrgan
Real-ESRGAN aims at developing Practical Algorithms for General Image Restoration
Decision gist · record as of 2026-08-14
Yes, if you need practical image or video upscaling. Real-ESRGAN is well-established with no known vulnerabilities, permissively licensed, and requires only standard Python dependencies. The dormant maintenance status is not a blocker—the package is feature-complete and stable. Install it if you have a GPU or can tolerate CPU inference; expect to download model weights on first use.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires PyTorch >= 1.7; model weights are downloaded on first use and can be several hundred MB.
- Low friction: pure Python wheel with no compiled dependencies.
- Maintenance is dormant (last commit 2024-08-06, 1424 days since release), but the repo remains active with no known vulnerabilities.
License · maintenance · safety
BSD-3-Clause License (permissive) — BSD-3-Clause is permissive: you can use, modify, and distribute this package freely in commercial or private projects, provided you include the license notice and disclaim liability.
last release 2022-09-20 (1424 days) · last repo commit 2024-08-06 · 36,478 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 184,216 downloads/mo, #10,043 on PyPI
Alternatives
Verify before relying
pip install realesrgan
from realesrgan import RealESRGANer
upsampler = RealESRGANer(scale=4, model_name='RealESRGAN_x4plus')
output, _ = upsampler.enhance(input_img, outscale=4)- Whether pre-trained model weights are automatically cached or must be manually downloaded.
- Exact memory and disk requirements for different model variants (x2plus, x4plus, anime_6B, etc.).
- Performance characteristics (inference time, VRAM usage) on typical hardware.
- Supported input and output formats beyond what the fact sheet explicitly documents.
What it is and what it does
Real-ESRGAN is a practical image and video restoration tool that enlarges low-resolution content using deep neural networks. It extends ESRGAN with models trained purely on synthetic data, making it robust to real-world degradation without requiring paired high/low-resolution training data. The package includes general-purpose upscaling models and specialized anime video models, plus integration with GFPGAN for face enhancement.
You use it by instantiating an upsampler with a chosen model and calling enhance() on input images or frames. It handles grayscale, 16-bit, and alpha-channel images, supports arbitrary output scales via LANCZOS4 resizing, and offers a tile option for processing large images. The package depends on torch, torchvision, basicsr, gfpgan, facexlib, numpy, opencv-python, Pillow, and tqdm.
Use it for
- Upscale low-resolution photographs or screenshots by 2x to 4x without training custom models.
- Restore anime images and videos using the specialized anime models.
- Enhance faces in portraits or video frames by combining Real-ESRGAN with integrated GFPGAN.
- Process archived or degraded video footage frame-by-frame to improve visual quality.
- Batch upscale images with tile mode to handle memory constraints on large files.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need practical image or video upscaling.
Real-ESRGAN is well-established with no known vulnerabilities, permissively licensed, and requires only standard Python dependencies. The dormant maintenance status is not a blocker—the package is feature-complete and stable. Install it if you have a GPU or can tolerate CPU inference; expect to download model weights on first use.
Install
realesrgan on PyPI
Before you install
Low friction: pure Python wheel with no compiled dependencies. Maintenance is dormant (last commit 2024-08-06, 1424 days since release), but the repo remains active with no known vulnerabilities. Nine runtime dependencies include torch, torchvision, and opencv-python, which are substantial but standard for computer vision work.
Requires PyTorch >= 1.7; model weights are downloaded on first use and can be several hundred MB.
License in practice
BSD-3-Clause is permissive: you can use, modify, and distribute this package freely in commercial or private projects, provided you include the license notice and disclaim liability.
Quickstart
pip install realesrgan
from realesrgan import RealESRGANer
upsampler = RealESRGANer(scale=4, model_name='RealESRGAN_x4plus')
output, _ = upsampler.enhance(input_img, outscale=4)Verify before relying
- Whether pre-trained model weights are automatically cached or must be manually downloaded.
- Exact memory and disk requirements for different model variants (x2plus, x4plus, anime_6B, etc.).
- Performance characteristics (inference time, VRAM usage) on typical hardware.
- Supported input and output formats beyond what the fact sheet explicitly documents.
Package facts
| License | BSD-3-Clause License permissive |
| Python support | Not specified |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 9 packagesbasicsrfacexlibgfpgannumpyopencv-pythonPillowtorchtorchvisiontqdm |
| Maintenance | Dormant 1,424 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 184,216 / month, #10,043 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Development Status :: 4 - BetaLicense :: OSI Approved :: Apache Software LicenseOperating System :: OS IndependentProgramming Language :: Python :: 3Programming Language :: Python :: 3.7Programming Language :: Python :: 3.8 |
Evidence: realesrgan-0.3.0-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “image upscaling super-resolution”
- realesrganReal-ESRGAN upscales images and videos using deep learning models…
- gfpganGFPGAN restores and enhances degraded faces in images using a…
- torchsrProvides pretrained neural network models for image super-resolution…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also basicsr · gfpgan · rembg · ncnn · resize-right · facexlib · icevision · segmentation-models-pytorch · nvidia-modelopt · runware