# fallenshock/FlowEdit

Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"

Repository: https://github.com/fallenshock/FlowEdit
Canonical: https://ross.abutalabs.com/products/flowedit
Homepage: https://matankleiner.github.io/flowedit/
Language: Python
License: MIT
License Family: permissive
Topics: diffusion-models, editing, editing-image, generative-model, rectified-flow, text-guided-image-editing, text-guided-image-manipulation
Last push: 2026-08-16T14:52:19+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 35, longevity 46
- inputs: {"age_days": 649, "days_push": 17, "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 1014, forks 51 (observed 2026-08-28T04:03:13.940006+00:00)

## What it is
Official PyTorch implementation of FlowEdit, an ICCV 2025 method for inversion-free, text-based editing of real images using pre-trained flow/diffusion models like Stable Diffusion 3 and FLUX. It constructs a direct ODE between source and target prompt distributions, avoiding image inversion and working model-agnostically.

## Use cases
- edit real images with a text prompt
- inversion-free image editing with diffusion models
- edit photos using Stable Diffusion 3 or FLUX
- research code for text-guided image manipulation
- zero-shot image editing without optimization
- use FlowEdit in ComfyUI workflows

## When to choose
- you need text-based editing of real images without inversion
- you want a model-agnostic editing method that transfers across flow model architectures
- you want state-of-the-art editing results with SD3 or FLUX
- you want to reproduce or build on published research

## When to avoid
- you need general-purpose image editing with masks or inpainting controls
- you lack a CUDA GPU or the required VRAM for SD3/FLUX
- you need a polished GUI application rather than scripts and YAML configs
- you need video editing as a first-class feature

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning
- domain: image-processing, artificial-intelligence, computer-vision
- platform: python, cross-platform
- tags: diffusion-models, rectified-flow, text-guided-image-editing, stable-diffusion-3, flux, inversion-free, pytorch, research-paper, comfyui, gpu, linux

## Member repositories
- fallenshock/FlowEdit (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:13.940006+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 2026-08-30T07:11:32.731157+00:00, confidence not recorded.
  - readme: https://github.com/fallenshock/FlowEdit (fetched 2026-08-28T04:03:13.940006+00:00, sha 38372ed8bd75)
  - homepage: https://matankleiner.github.io/flowedit/ (fetched 2026-08-29T13:10:53.775230+00:00, sha 922c7e6e364d)
- Data as of 2026-08-30T08:39:29.467469+00:00.