# instantX-research/InstantID

InstantID: Zero-shot Identity-Preserving Generation in Seconds 🔥

Repository: https://github.com/instantX-research/InstantID
Canonical: https://ross.abutalabs.com/products/instantid
Homepage: https://instantid.github.io/
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-07-18T04:12:48+00:00

## Health v2 (maintenance only)
Score: 26/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 71
- inputs: {"age_days": 996, "days_push": 776, "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 11987, forks 885 (observed 2026-08-28T04:10:50.608172+00:00)

## What it is
InstantID is a tuning-free, zero-shot identity-preserving image generation method built on diffusion models, generating customized images in various styles from a single reference face image. It works as a plug-and-play module with pre-trained text-to-image models like SD1.5 and SDXL via a novel IdentityNet and has been merged into Hugging Face diffusers.

## Use cases
- generate images of a person in different poses and styles from one photo
- create identity-preserving avatars without fine-tuning
- personalize stable diffusion outputs with a single face image
- avoid DreamBooth or LoRA training for face customization
- integrate face-consistent generation into an SDXL pipeline

## When to choose
- you need high face fidelity from only a single reference image
- you want tuning-free personalization with no per-identity training
- you already use SD1.5/SDXL or diffusers and want a plug-and-play identity module

## When to avoid
- you need style transfer rather than identity preservation
- you lack a GPU or cannot run diffusion model inference
- you need to preserve non-face subjects like objects or pets

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning, stable-diffusion
- domain: artificial-intelligence, image-processing, computer-vision, deep-learning
- platform: python, cross-platform
- tags: identity-preservation, text-to-image, diffusion-models, zero-shot, face-generation, controlnet, sdxl, tuning-free, gpu

## Member repositories
- instantX-research/InstantID (main) score 26

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:50.608172+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 2026-08-29T17:15:01.494020+00:00, confidence not recorded.
  - readme: https://github.com/instantX-research/InstantID (fetched 2026-08-28T04:10:50.608172+00:00, sha 97a44e444934)
  - homepage: https://instantid.github.io/ (fetched 2026-08-29T08:12:43.427735+00:00, sha c2a4e05e88ee)
- Data as of 2026-08-30T08:39:29.467469+00:00.