# W2GenAI-Lab/LucidFlux

LucidFlux: Caption-Free Photo-Realistic Image Restoration via a Large-Scale Diffusion Transformer, ICLR 2026

Repository: https://github.com/W2GenAI-Lab/LucidFlux
Canonical: https://ross.abutalabs.com/products/lucidflux
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-05-26T15:12:17+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 84, release rhythm 35, longevity 25
- inputs: {"age_days": 363, "days_push": 99, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1293, forks 111 (observed 2026-08-28T04:04:16.139020+00:00)

## What it is
LucidFlux is a caption-free photo-realistic image restoration model built on a large-scale diffusion transformer, released with inference and training code in Python. It supports restoring degraded photos without needing input captions and has been extended to 4K image restoration.

## Use cases
- restore old degraded photos to photo-realistic quality
- super-resolve low-quality images without writing captions
- run 4K image restoration with a diffusion transformer
- train a custom image restoration model
- integrate image restoration into an API via fal.ai

## When to choose
- you need high-fidelity blind image restoration without caption prompts
- you want a research-grade diffusion-based restoration model with training code
- you need 4K-capable photo restoration on GPU hardware

## When to avoid
- you need lightweight restoration on CPU or edge devices
- you need a simple classical denoising filter rather than generative restoration
- you require a permissive license for commercial use without review

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning, stable-diffusion
- domain: image-processing, artificial-intelligence, deep-learning
- platform: python
- tags: image-restoration, diffusion-transformer, super-resolution, photo-restoration, caption-free, iclr-2026, gpu, linux

## Member repositories
- W2GenAI-Lab/LucidFlux (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:16.139020+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 2026-08-30T04:54:08.252724+00:00, confidence not recorded.
  - readme: https://github.com/W2GenAI-Lab/LucidFlux (fetched 2026-08-28T04:04:16.139020+00:00, sha eea99fcdb6ff)
- Data as of 2026-08-30T08:39:29.467469+00:00.