# magic-research/magic-edit

MagicEdit: High-Fidelity Temporally Coherent Video Editing

Repository: https://github.com/magic-research/magic-edit
Canonical: https://ross.abutalabs.com/products/magic-edit
Homepage: https://magic-edit.github.io/
License: BSD-3-Clause
License Family: permissive
Archived: true
Last push: 2023-08-29T04:28:12+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 78
- inputs: {"age_days": 1105, "days_push": 1100, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1790, forks 100 (observed 2026-08-28T04:05:36.966740+00:00)

## What it is
MagicEdit is a research implementation of a diffusion-based video editing model from ByteDance that disentangles appearance and motion for high-fidelity, temporally coherent edits. It supports video stylization, local editing, video-MagicMix, and video outpainting driven by text prompts.

## Use cases
- edit videos with text prompts
- stylize a video into cartoon or another style
- replace subjects in a video like dog to cat
- outpaint video frames to extend the scene
- perform local edits on video while preserving motion

## When to choose
- you need research-grade text-driven video editing with temporal coherence
- you want to experiment with or build on a published video editing model
- you have GPU resources and want to run video stylization or outpainting locally

## When to avoid
- you need a polished end-user video editor with a GUI
- you lack a GPU or ML deployment experience
- you need production-supported software with frequent updates

## Facets
- artifact type: library
- maturity: experimental
- function: video-processing, image-processing, machine-learning, deep-learning
- domain: computer-vision, artificial-intelligence, image-processing
- platform: python
- tags: video-editing, diffusion-models, text-guided-editing, research-code, video-stylization, temporally-coherent, video, gpu, linux

## Member repositories
- magic-research/magic-edit (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:36.966740+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 2026-08-30T03:23:17.230147+00:00, confidence not recorded.
  - readme: https://github.com/magic-research/magic-edit (fetched 2026-08-28T04:05:36.966740+00:00, sha 1b8b06efd590)
  - homepage: https://magic-edit.github.io/ (fetched 2026-08-29T11:02:24.174694+00:00, sha af1abcf68529)
- Data as of 2026-08-30T08:39:29.467469+00:00.