#4949th of 56 in Image and videoFor image, video and design, image and design
waifu2x-ncnn-vulkan
Command-line waifu2x image upscaler and denoiser using ncnn and Vulkan
- Stars
- 3.5k
- License
- MIT
- Last commit
- Feb 2026
- Last release
- Sep 2025
- Language
- C++
3,347 on 21 Apr 20263,502 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Upscales and denoises anime-style images with the waifu2x models, running inference through ncnn on any Vulkan-capable GPU (Intel, AMD, NVIDIA, Apple Silicon). Ships as a portable Windows/Linux/macOS executable with models bundled, so no CUDA or Caffe install is needed. It takes jpg/png/webp files or directories and supports tiling, multi-GPU and TTA mode.
Who it is for: Users upscaling anime-style images in batch from the command line on any GPU
Strengths
- Portable binary with models included; no CUDA or Caffe runtime required
- Runs on Intel, AMD, NVIDIA and Apple Silicon GPUs through Vulkan
- README benchmarks show faster times than waifu2x-caffe-cui on images up to 2000x2000
- Tile size and load:proc:save thread counts can be tuned to limit GPU memory
Weaknesses
- CLI only; no GUI or web interface
- Slower than waifu2x-caffe-cui at 4000x4000 input in the README benchmarks
- Scope is waifu2x models only; no other upscalers or video pipeline
- Driver crashes are possible; README advises updating the GPU driver
What it needs
- GPU optional
- Needs Vulkan driver, ncnn, libwebp, libjpeg-turbo, libpng
- Models: waifu2x cunet, upconv_7_anime_style_art_rgb, jpg, png, webp
Quick answers
Is waifu2x-ncnn-vulkan open source?
Yes. Its code is on GitHub under the MIT license.
Does waifu2x-ncnn-vulkan need a GPU?
No. It runs on a CPU and can use a GPU when you have one.
Also in Image and video
See all 56| Rank | Project | Score |
|---|---|---|
| 1 | frigateLocal NVR with realtime object detection for IP cameras | 92 out of 100 |
| 2 | MoneyPrinterTurboGenerates short videos from a topic, with script, footage, voiceover and subtitles | 88 out of 100 |
| 3 | toonflow-appInfinite-canvas tool for AI short drama and comic video production | 86 out of 100 |
| #4 | ComfyUINode-graph editor for running diffusion image, video, audio and 3D workflows | 85 out of 100 |
| #5 | InvokeAIWeb UI for diffusion image generation with inpainting canvas and node workflows | 85 out of 100 |