ESRGAN x4 INT8

Cloud CPUAWS Graviton G4

INT8-quantized 4× single-image super-resolution models for enhancing low-resolution RGB images, especially urban scenes with fine repetitive details. Built on a 23-block RRDBNet backbone, they use tiled 128×128 inputs and produce 512×512 outputs per tile, balancing image fidelity with reduced model size and faster inference on ARM-oriented runtimes.View on Hugging Face

Accuracyvs base model

PSNR26.81 dB↓0.22 dB

SSIM0.81↓0.01

Frames / sec

About this version

Original Model:: RRDB_PSNR_x4 by Not specified — Repository
Optimization/conversion:: Arm-Optimized version for execution on Arm-based platforms.
Converted/optimized by:: Arm
License:: The Original Model and the Optimized Model are subject to Apache-2.0.

This repository contains a converted or optimized version of the Original Model (the “Optimized Model”). The Original Model has been converted or optimized as described above for execution on Arm-based platforms.

No retraining or fine-tuning of the Original Model was performed as part of the conversion or optimization. The conversion or optimization was not intended to change the Original Model’s behavior or intended use.

Original Model and Documentation

For information about the Original Model, including its development, training data, intended uses, limitations and other relevant information, please refer to the Original Model repository. Information in that repository was provided by the original developer or other third parties and, unless expressly stated otherwise, has not been independently verified by Arm.

Licenses and Third-Party Terms

Use of the Original Model and the Optimized Model is subject to the applicable licenses, usage restrictions and other terms identified above and in the relevant repositories. Publication of the Optimized Model does not grant any rights beyond those provided under the applicable license terms.

You are responsible for reviewing those terms and ensuring that your use of the Original Model and the Optimized Model is permitted.

About this version

Original Model:: RRDB_PSNR_x4 by Not specified — Repository
Optimization/conversion:: Arm-Optimized version for execution on Arm-based platforms.
Converted/optimized by:: Arm
License:: The Original Model and the Optimized Model are subject to Apache-2.0.

Original Model and Documentation

Licenses and Third-Party Terms

You are responsible for reviewing those terms and ensuring that your use of the Original Model and the Optimized Model is permitted.

Metric	Resultsvs baseline	Change
End to end latency p50	704.42ms1762.73ms	−1058.31 ms
End to end latency p90	779.70ms1764.72ms	−985.02 ms
End to end latency p99	819.72ms1771.11ms	−951.39 ms
Peak memory	795.07MB799.74MB	−4.67 MB
Average memory	754.93MB797.35MB	−42.42 MB
Time to first inference	906.88ms1968.65ms	−1061.77 ms
Frames per second	1.42fps0.57fps	+0.85 fps
PSNR	26.81dB27.03dB	−0.22 dB
SSIM	0.810.82	−0.01

Dataset	urban100	urban100
Sample count	100	100
Batch size	1	1
Input resolution	128x128x3	128x128x3
Runs	100	100
Warmup runs	10	10

ESRGAN x4 INT8

ESRGAN x4 INT8

Model Evaluation

Configuration

Comparability

Purpose of this Release