# yeates/PromptFix

[NeurIPS 24] PromptFix: You Prompt and We Fix the Photo

Repository: https://github.com/yeates/PromptFix
Canonical: https://ross.abutalabs.com/products/promptfix
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-10-04T21:47:11+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 59
- inputs: {"age_days": 831, "days_push": 698, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1070, forks 60 (observed 2026-08-28T04:03:28.048249+00:00)

## What it is
PromptFix is a PyTorch implementation of a diffusion-model-based image restoration model that follows natural language instructions to fix degraded photos. It supports tasks like colorization, dehazing, deblurring, object and watermark removal, and low-light enhancement, and includes pre-trained weights, training code, and a 1M+ sample dataset.

## Use cases
- remove watermark from a photo with a text instruction
- deblur old blurry photos
- colorize black and white images
- enhance low-light photos
- remove unwanted objects from images
- dehaze outdoor photos
- train an instruction-based image restoration model

## When to choose
- you need instruction-driven image restoration across many degradation types
- you want pre-trained weights and a large paired training dataset
- you need structure-preserving photo enhancement with a diffusion backbone

## When to avoid
- you need real-time or on-device processing with limited GPU resources
- you only need a single fixed restoration task where a specialized model is simpler
- you cannot run diffusion inference locally

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: python
- tags: diffusion-models, image-restoration, instruction-following, pytorch, neurips-2024, photo-enhancement, linux, gpu

## Member repositories
- yeates/PromptFix (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:28.048249+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 2026-08-30T06:54:14.813665+00:00, confidence not recorded.
  - readme: https://github.com/yeates/PromptFix (fetched 2026-08-28T04:03:28.048249+00:00, sha 8b1bdf910f0a)
- Data as of 2026-08-30T08:39:29.467469+00:00.