Skip to content
RepoQuarry

QualityScaler Djdefrag/QualityScaler

Windows app that upscales and denoises photos and videos with ONNX models on any DirectX 12 GPU.

QualityScaler at a glance

Checked 16 Oct 2026

QualityScaler is a Python desktop app for Windows that upscales and denoises images and videos with models from BSRGAN, Real-ESRGAN and IRCNN, running through onnxruntime-directml on AMD, Intel or Nvidia GPUs with automatic tiling. Code is MIT. GitHub has no prebuilt binaries: the packaged app is sold on itch.io and Steam, and the free route is running from source with model files downloaded separately; their licence is not stated. Windows only; no face recovery. Active: release 2026.4 (Aug 2026).

Cite this pageRepoQuarry. “QualityScaler.” Checked 16 Oct 2026. https://repoquarry.dev/projects/qualityscaler

How much of Topaz Gigapixel does it replace?

3 of 6 jobs. Upscale and denoise many files on Windows; no face recovery, generative models or plugins.

Partial: Photo upscaling up to 16x Partial replacement
No: Face recovery on blurry or low-res faces Not covered
Partial: Recover and redefine low-quality or compressed images Partial replacement
No: Upscaling art Not covered
Partial: Batch upscaling of many images Partial replacement
No: Photoshop and Lightroom plugin Not covered

For face recovery on blurry or low-res faces, upscaling art, Photoshop and Lightroom plugin, see other Topaz Gigapixel alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction
QuestionAnswerSourceChecked
Runs on my machine? Confirmed: Yes, fully offline Privacy focused - no internet connection required, everything runs on your PC github.com · QualityScaler#-features ↗ today
Code licence Confirmed: MIT github.com · LICENSE ↗ today
Model weights licence Unknown: README links a separate AI-model archive, while LICENSE/LICENCE notices cover software and dependencies; no complete per-weight grant was verified for the downloadable ONNX models github.com · QualityScaler#readme ↗ today
Commercial use Unknown: Application code is MIT; commercial-use terms of the separately downloaded model weights remain unverified in the reviewed README and licence notices github.com · LICENSE.txt ↗ today
Paid services needed? Partial: Source is free; the packaged app is sold on itch.io and Steam github.com · QualityScaler#readme ↗ today
Hardware Confirmed: 8GB RAM; any DirectX 12 GPU with 4GB VRAM or more Any DirectX12 compatible GPU with ≥ 4GB VRAM github.com · QualityScaler#-requirements ↗ today
Operating systems Partial: Windows 10 and 11 only Windows 10 / Windows 11 github.com · QualityScaler#-requirements ↗ today
Setup effort Confirmed: Medium for the free route, run from source in Python and add models and ffmpeg yourself Extract the AI models files in the /AI-onnx folder github.com · QualityScaler#-make-it-work-by-yourself- ↗ today

Get the weekly top 5 and what’s new, every Monday

We’ll email you once a week. Unsubscribe any time. We never share your email.

Cookie preferences

Choose what we may store in your browser. You can change this any time from Cookie settings in the footer.

Essential

Security, and remembering this choice. rq-consent · 12 months

Always on