# chaiNNer

> chaiNNer chains image operations as nodes: load a folder or video, upscale with a model, then resize, sharpen or save. It loads community models (e.g. from OpenModelDB) through its Spandrel library, which also covers GFPGAN and CodeFormer face restoration. It ships no default model, so the weights licence is whatever you load. Code is GPL-3.0. It supports Nvidia, AMD, Intel and Apple GPUs, with CPU fallback. It is not a one-click tool; you build the chain yourself. Last release v0.25.1 (Oct 2025); commits continue.

- Repository: https://github.com/chaiNNer-org/chaiNNer
- Official docs: https://github.com/chaiNNer-org/chaiNNer/wiki
- Checked: 2026-10-16

## What it replaces
- Topaz Gigapixel: Photo upscaling up to 16x, Face recovery on blurry or low-res faces, Upscaling art, Batch upscaling of many images (partial). Covers these jobs with models you choose and chains you build; no generative upscaling models or Photoshop plugin.

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Yes, desktop app with its own integrated Python | https://github.com/chaiNNer-org/chaiNNer#installation | 2026-10-16 |
| Code licence | Partial | GPL-3.0 (copyleft) | https://github.com/chaiNNer-org/chaiNNer/blob/main/LICENSE | 2026-10-16 |
| Model weights licence | Partial | No default model; licence depends on the model you load (many community models are non-commercial) | https://github.com/chaiNNer-org/spandrel#readme | 2026-10-16 |
| Commercial use | Partial | GPL-3.0 code; commercial use depends on each model's licence | https://github.com/chaiNNer-org/spandrel#readme | 2026-10-16 |
| Paid services needed? | Confirmed | None required | https://github.com/chaiNNer-org/chaiNNer#readme | 2026-10-16 |
| Hardware | Confirmed | Nvidia (CUDA/TensorRT), AMD (NCNN, ROCm on Linux), Intel (NCNN), Apple Silicon (MPS); CPU fallback | https://github.com/chaiNNer-org/chaiNNer#gpu-support | 2026-10-16 |
| Operating systems | Confirmed | Windows, macOS, Linux | https://github.com/chaiNNer-org/chaiNNer#readme | 2026-10-16 |
| Setup effort | Confirmed | Easy installer; then pick a framework in the Dependency Manager | https://github.com/chaiNNer-org/chaiNNer#installation | 2026-10-16 |

GitHub: 6076 stars, 377 forks, last push 2026-10-11, archived: no.
---
Source: RepoQuarry (https://repoquarry.dev/projects/chainner). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.