# bloc97/Anime4K

A High-Quality Real Time Upscaler for Anime Video

Repository: https://github.com/bloc97/Anime4K
Canonical: https://ross.abutalabs.com/products/anime4k
Homepage: https://bloc97.github.io/Anime4K/
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: anime, upscaling, upsampling, video-processing, video, anime4k, computer-graphics, cnn, neural-networks, convolutional-neural-networks, super-resolution, denoising-algorithms, anime-upscaling
Last push: 2024-08-17T12:09:12+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2577, "days_push": 746, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 21295, forks 1397 (observed 2026-08-28T04:11:31.525061+00:00)

## What it is
Anime4K is a set of open-source, high-quality real-time anime upscaling and denoising algorithms implemented as GLSL shaders, primarily for use with video players like MPV. It upscales native 1080p anime to 4K in real time using fast CNN-based techniques without re-encoding the source video.

## Use cases
- upscale anime to 4k in real time while watching
- denoise anime video during playback
- shaders to make 1080p anime look sharp on a 4k monitor
- real-time super resolution for anime without re-encoding
- improve anime video quality in mpv
- upscale anime without losing quality or disk space

## When to choose
- you watch native 1080p anime (h.264/h.265/VC-1) on a 4K display and want real-time upscaling
- you want a lightweight shader-based solution in MPV, Plex, or IINA rather than offline batch processing
- you need speed and simplicity over maximum quality on heavily degraded sources

## When to avoid
- your source is low-resolution or old content (720p, 480p, DVDs, pre-digital anime) with heavy artifacts
- you need offline batch upscaling of images or heavily degraded video, where SRGAN tools like Real-ESRGAN or waifu2x perform better
- you don't have a GPU or a player that supports GLSL shaders

## Facets
- artifact type: library
- maturity: stable
- function: video-processing, image-processing, machine-learning, deep-learning
- domain: image-processing, computer-vision, media
- platform: cross-platform, windows
- tags: anime, upscaling, super-resolution, denoising, glsl-shaders, mpv, real-time, cnn, video, linux, macos, gpu

## Member repositories
- bloc97/Anime4K (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:31.525061+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:58:26.436972+00:00, confidence not recorded.
  - readme: https://github.com/bloc97/Anime4K (fetched 2026-08-28T04:11:31.525061+00:00, sha 8aa469d7c2f4)
  - homepage: https://bloc97.github.io/Anime4K/ (fetched 2026-08-29T07:56:51.991901+00:00, sha aa008275d116)
- Data as of 2026-08-30T08:39:29.467469+00:00.