Skip to main content

chaiNNer

Free, open-source desktop app that chains AI upscaling models and other image steps into node-based batch workflows.

Try free
Rate it

Is chaiNNer free?

Yes, chaiNNer is free and open source; you run it yourself. What you get: Open source (GPLv3): free desktop app, nothing gated behind donations. Watch out: not supported on macOS 10.x or Windows 8.1 and older. What it does: free, open-source desktop app that chains AI upscaling models and other image steps into node-based batch workflows. Verified on chainner.app.

Free plan at a glance

Includes
  • Open source (GPLv3): free desktop app, nothing gated behind donations

No usage limit: free and open source, and nothing is gated behind donations.

Duration
Free forever

How to get it free

Self-host
  1. Download the latest release from GitHub and run the installer for your system
  2. Open chaiNNer and let it set up its own Python environment
  3. Install PyTorch, NCNN, ONNX or TensorRT from the Dependency Manager, then build a load, upscale and save chain

Good to know

  • Not supported on macOS 10.x or Windows 8.1 and older
  • You install a neural network framework from the Dependency Manager before you can upscale
Get the free plan

github.com, Checked Source

Overview

chaiNNer is a GPLv3 node editor for Windows, macOS and Linux. It began as an AI upscaling app and runs community super-resolution models through PyTorch, NCNN, ONNX or TensorRT, alongside background removal, color work, batch runs and video frames.

Key features

  • AI upscaling with ESRGAN, SPAN, OmniSR and dozens of other model architectures
  • Face restoration and color-fix nodes for upscaled images
  • Background removal with ONNX models
  • Batch runs over folders of images, video frames or a folder of models
  • Condition, math and text nodes for building logic without code
  • Saved chains can run from the command line

Best for

People
People who upscale photos, artwork or game textures with AI models, Artists working with pixel art, textures and normal maps, Anyone who wants repeatable batch image pipelines without writing scripts

Limitations

  • Supports a limited set of neural network architectures for now
  • On Apple Silicon, ONNX runs on the CPU only and NCNN may not work on some setups
  • Some NCNN users with non-Nvidia GPUs may get all-black outputs and need to set tiling by hand

Screenshot

chaiNNer screenshot, as published by chaiNNer
Image published by chaiNNer

FAQ

Is chaiNNer free?
Yes. chaiNNer is free and open source. That's from chainner.app, checked on November 16, 2026.
What does the chaiNNer free plan include?
No usage limit: free and open source, and nothing is gated behind donations. We read this on chainner.app on November 16, 2026.
How do I get chaiNNer for free?
Download the latest release from GitHub and run the installer for your system. Then open chaiNNer and let it set up its own Python environment. Finally, install PyTorch, NCNN, ONNX or TensorRT from the Dependency Manager, then build a load, upscale and save chain. Start at github.com/chaiNNer-org/chaiNNer. We last checked these steps on November 16, 2026.
Do I need an account to use chaiNNer for free?
Not for the free option: the free way to use chaiNNer is to host it yourself. To start, download the latest release from GitHub and run the installer for your system (checked November 16, 2026).
What's the catch with chaiNNer's free plan?
Two things to know. Not supported on macOS 10.x or Windows 8.1 and older. You install a neural network framework from the Dependency Manager before you can upscale. We last checked this on November 16, 2026.
What are free alternatives to chaiNNer?
Upscale.media, Clipdrop and Upscayl all have free plans. We checked each of their free offers on November 16, 2026.
Do I need to install Python first?
No. chaiNNer downloads an isolated Python build on startup and installs other dependencies through its Dependency Manager. You can use your system Python instead (3.10 or later).Source: github.com/chaiNNer-org/chaiNNer
Which GPUs does chaiNNer support?
Nvidia through PyTorch (CUDA), ONNX and TensorRT; AMD through NCNN, or PyTorch with ROCm on Linux; Apple Silicon through PyTorch MPS; and Intel through NCNN. All frameworks can fall back to the CPU.Source: github.com/chaiNNer-org/chaiNNer
Where do I get upscaling models?
Most of the models people use come from OpenModelDB, a collection of community-trained super-resolution models.Source: chainner.app

Free alternatives

More free AI Image Generation tools

All free AI Image Generation tools

Sources

Free offer
Verified on chainner.app
…nodes onto a canvas, connect them, and press run. Free and open source, for Windows, macOS, and…
Page content

Checked against

How we check free offers

Added by our team. We check that a free option exists and record the date; signing up is between you and chaiNNer, so read their privacy policy and terms first.

Is this your tool? Claim it · Promote it