Real-ESRGAN (fastforks)

Real-ESRGAN (Wang et al., ICCVW 2021): practical blind super-resolution. ESRGAN's RRDBNet (23 residual-in-residual dense blocks of 3 x 3 convs, then two nearest x2 upsamplings with convs) trained with a GAN on synthetic real-world degradations (blur, noise, JPEG, two-stage). realesrgan-x4plus is the general x4 model, realesrgan-x2plus the x2 one (the input pixel-unshuffled first), realesrgan-x4plus-anime-6b a 6-block x4 model for anime images.

This repo is the model card only: it holds no weights. fastforks downloads the authors' weights from the links below and checks them against the SHA-256 in the model card. Code: https://github.com/kleinicke/fastforks. Kernels for the kernels backend: https://huggingface.co/fastforks/kernels. Original code: https://github.com/xinntao/Real-ESRGAN at a4abfb2979a7.

Task: restoration. Tier: experimental. Outputs: image. Backends: portable, kernels.

Input: one RGB image. Output: the restored (upscaled) image, model.scale x the input size, RGB in [0, 1]. tile= runs large images in overlapping tiles (as ComfyUI).

Variants

Weights are the authors' files (not hosted here). SHA-256 shortened to 12 hex digits; the full hashes are in card.toml.

variant weights (author link) SHA-256 weights licence commercial use
realesrgan-x2plus@1 RealESRGAN_x2plus.pth 49fafd45f8fd BSD-3-Clause yes
realesrgan-x4plus@1 RealESRGAN_x4plus.pth 4fa0d38905f7 BSD-3-Clause yes
realesrgan-x4plus-anime-6b@1 RealESRGAN_x4plus_anime_6B.pth f872d837d3c9 BSD-3-Clause yes

Licence

Original-code strings, recorded as they are (research-use package; check them before any other use). Code: BSD-3-Clause. Licence files: LICENSE, LICENSE-BasicSR.

Training data: DIV2K (NTIRE 2017 super-resolution, 1000 2K images), Flickr2K (2650 2K Flickr images, EDSR), OutdoorSceneTraining (OST, SFTGAN) (docs/Training.md: DF2K (DIV2K + Flickr2K) + OST, HR images only; the anime model: anime images (not named).).

Licence evidence: https://github.com/xinntao/Real-ESRGAN/blob/a4abfb2979a7bbff3f69f58f58ae324608821e27/LICENSE ; https://github.com/xinntao/Real-ESRGAN/releases ; https://github.com/XPixelGroup/BasicSR/blob/8d56e3a045f9fb3e1d8872f92ee4a4f07f886b0a/LICENSE.txt

Input size

any H x W >= 32, padded to a multiple of 32 Tested up to (portable, fp16, RTX 5060 Ti 16 GB): realesrgan-x2plus 1988x2880 (11.9 GiB peak allocated, 14.6 reserved); realesrgan-x4plus 1024x2048 (12.7 GiB peak allocated, 14.9 reserved); realesrgan-x4plus-anime-6b 1024x2048 (12.7 GiB peak allocated, 14.9 reserved)

Speed and memory

Restoration, synthetic, 512x512

model time (ms): original portable kernels kernels speed-up: vs original vs portable peak memory (MiB, allocated / reserved): original portable kernels
realesrgan-x2plus 128 97.0 80.2 1.6x 1.2x 592 / 656 333 / 394 333 / 394
realesrgan-x4plus 631 471 403 1.6x 1.2x 2266 / 2496 1220 / 1454 1220 / 1434
realesrgan-x4plus-anime-6b 205 155 132 1.6x 1.2x 2243 / 2472 1197 / 1430 1197 / 1410

Startup and first call

model load (s): original portable kernels first call (s): original portable kernels
realesrgan-x2plus 1.54 0.25 1.26 0.42 0.26 0.11
realesrgan-x4plus 1.62 0.20 2.43 1.64 0.62 0.43
realesrgan-x4plus-anime-6b 1.30 0.10 1.30 0.50 0.23 0.15

torch.compile and TensorRT

model input time: portable compile TensorRT fp16 kernels preparation: compile (cold first call) ONNX + engine build peak MiB (alloc / reserved): portable compile kernels TensorRT (engine + torch)
realesrgan-x2plus 512x512 97.0 ms 85.2 ms 59.7 ms 80.2 ms 33.2 s 40.3 s 333 / 394 323 / 394 333 / 394 476 / 500
realesrgan-x4plus 512x512 471 ms 354 ms 273 ms 403 ms 29.9 s 55.2 s 1220 / 1454 1091 / 1394 1220 / 1434 1546 / 1594
realesrgan-x4plus-anime-6b 512x512 155 ms 114 ms 83.1 ms 132 ms 10.0 s 36.7 s 1197 / 1430 1068 / 1370 1197 / 1410 1505 / 1552

Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB).

Hardware, protocol and caveats: benchmarks (RTX 5060 Ti 16 GB, sm_120). Original = the original code named above at the precision of portable (fp16, or (fp32) where fp16 is unsupported or not finite); original fp32 = its default fp32 run. Peak memory is the torch allocator's.

Accuracy

On public datasets

DIV2K validation HR images and the official bicubic LR images (x4 or x2, the model's scale), one pass without tiles; PSNR and SSIM (Gaussian 11 x 11, sigma 1.5) on the Y channel (BT.601) of the output rounded to uint8, scale px border cropped, averaged over the images. Real-ESRGAN is trained for real-world degradations, not bicubic, so its PSNR is low: the numbers compare our modes with the original, not with PSNR-oriented upscalers. Original = the original code named under Details, same checkpoint and protocol.

variant dataset original fp32 original fp16 portable kernels note
realesrgan-x4plus DIV2K validation (100 images), bicubic x4 LR -> HR, PSNR / SSIM (Y) 26.64 / 0.7580 26.64 / 0.7580 26.64 / 0.7580 26.64 / 0.7580
realesrgan-x2plus DIV2K validation (100 images), bicubic x2 LR -> HR, PSNR / SSIM (Y) 30.03 / 0.8592 30.03 / 0.8592 30.03 / 0.8592 30.03 / 0.8591
realesrgan-x4plus-anime-6b DIV2K validation (100 images), bicubic x4 LR -> HR, PSNR / SSIM (Y) 25.66 / 0.7357 25.66 / 0.7357 25.66 / 0.7357 25.66 / 0.7356

Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 45a5c91f, 2026-11-13.

Parity with the original code

Every variant is checked against reference outputs of the original code at its pinned commit. CPU, float32: max |diff| <= 1e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.02 and p99 <= 0.25 (absolute error of the output image).

Usage

from fastforks import restoration

model = restoration.load("realesrgan-x2plus")  # mode="auto": the preferred backend from recorded measurements
result = model("photo.png")  # tile=512 for large images
result.image  # [3, sH, sW] float32 RGB in [0, 1], s = model.scale
result.backend  # what ran, e.g. Backend(mode='portable', device='cuda:0', dtype='float16')

Install with pip install fastforks (see https://github.com/kleinicke/fastforks). Other variants: realesrgan-x2plus, realesrgan-x4plus, realesrgan-x4plus-anime-6b.

Kernels backend

restoration.load("realesrgan-x2plus", mode="kernels") runs our CUDA kernels (cudnn_hacc). It needs a prebuilt kernel for your GPU architecture (downloaded from https://huggingface.co/fastforks/kernels on first use) and a recent NVIDIA driver; if there is none it raises BackendUnavailableError with the reason and never falls back silently. mode="auto" picks kernels when it can run and logs the choice.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Collection including fastforks/realesrgan