# orpatashnik/StyleCLIP

Official Implementation for "StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery" (ICCV 2021 Oral)

Repository: https://github.com/orpatashnik/StyleCLIP
Canonical: https://ross.abutalabs.com/products/styleclip
Language: HTML
License: MIT
License Family: permissive
Last push: 2023-05-30T19:57:17+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2026, "days_push": 1191, "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 4121, forks 564 (observed 2026-08-28T04:08:35.969929+00:00)

## What it is
Official implementation of StyleCLIP, a method for text-driven manipulation of StyleGAN-generated imagery using CLIP. It provides three approaches: CLIP-based latent optimization, a latent mapper network, and global style-space directions.

## Use cases
- edit images with text prompts
- manipulate StyleGAN latent vectors using natural language
- change facial attributes of generated faces via text
- explore CLIP-guided image editing
- reproduce StyleCLIP paper results
- run text-to-face editing in Colab

## When to choose
- you want text-guided editing of StyleGAN images
- you need the reference implementation of the StyleCLIP paper
- you work with GAN latent space manipulation research

## When to avoid
- you need general-purpose image editing of arbitrary photos without a StyleGAN inversion step
- you want a production-ready maintained product
- you need modern diffusion-based text-to-image editing

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, image-processing, llm-inference
- domain: deep-learning, computer-vision, image-processing, artificial-intelligence
- platform: python, cross-platform
- tags: stylegan, clip, text-driven-image-editing, image-manipulation, generative-models, research-code, iccv-2021, gpu

## Member repositories
- orpatashnik/StyleCLIP (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:35.969929+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 2026-08-29T18:23:06.785397+00:00, confidence not recorded.
  - readme: https://github.com/orpatashnik/StyleCLIP (fetched 2026-08-28T04:08:35.969929+00:00, sha 646974e3d7d6)
- Data as of 2026-08-30T08:39:29.467469+00:00.