# JackAILab/ConsistentID

[TPAMI 2026] ConsistentID : Portrait Generation with Multimodal Fine-Grained Identity Preserving

Repository: https://github.com/JackAILab/ConsistentID
Canonical: https://ross.abutalabs.com/products/consistentid
Homepage: https://ssugarwh.github.io/consistentid.github.io/
Language: Python
License: MIT
License Family: permissive
Last push: 2026-01-02T07:14:54+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 60, release rhythm 35, longevity 62
- inputs: {"age_days": 870, "days_push": 243, "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 1026, forks 81 (observed 2026-08-28T04:03:16.847350+00:00)

## What it is
ConsistentID is a diffusion-based portrait generation model and toolkit that preserves facial identity from a single reference image using multimodal fine-grained facial prompts (FaceParsing + FaceID). It ships with training/evaluation code, SD1.5 and SDXL variants, and the 500k+ image FGID dataset.

## Use cases
- generate personalized portraits from a single face photo
- preserve identity consistency in AI-generated faces
- create consistent character portraits from text prompts
- use an identity adapter with community SDXL LoRA models
- train or evaluate identity-preserving portrait generation models
- access a large fine-grained facial portrait dataset

## When to choose
- you need high-fidelity identity preservation in generated portraits without LoRA training
- you want rapid per-identity customization within seconds from one image
- you need an adapter compatible with other base models and LoRA modules
- you want a research-grade implementation with dataset and evaluation benchmark

## When to avoid
- you need general-purpose image generation without identity constraints
- you lack a GPU or cannot host large diffusion model checkpoints
- you need production-grade commercial face generation with licensing guarantees beyond the model weights' terms
- you want lightweight on-device inference

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning, stable-diffusion, sdk
- domain: artificial-intelligence, image-processing, computer-vision, deep-learning, large-language-models
- platform: python
- tags: identity-preservation, portrait-generation, diffusion-models, face-parsing, sdxl, research-paper, tpami, fgid-dataset, zero-shot-customization, adapter, gpu, docker, web-server

## Member repositories
- JackAILab/ConsistentID (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:16.847350+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 2026-08-30T07:07:56.654594+00:00, confidence not recorded.
  - readme: https://github.com/JackAILab/ConsistentID (fetched 2026-08-28T04:03:16.847350+00:00, sha 51d9d9c2f772)
  - homepage: https://ssugarwh.github.io/consistentid.github.io/ (fetched 2026-08-29T13:08:22.727460+00:00, sha b7d64dc582e0)
- Data as of 2026-08-30T08:39:29.467469+00:00.