# Kwai-Kolors/Kolors

Kolors Team

Repository: https://github.com/Kwai-Kolors/Kolors
Canonical: https://ross.abutalabs.com/products/kolors
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-11-13T05:17:23+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 56
- inputs: {"age_days": 789, "days_push": 658, "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 4615, forks 357 (observed 2026-08-28T04:08:55.109653+00:00)

## What it is
Kolors is a large-scale latent diffusion model for photorealistic text-to-image synthesis, trained with bilingual (Chinese and English) text understanding. The repository provides inference code, pretrained weights, and extensions including ControlNet, IP-Adapter, inpainting, and DreamBooth-LoRA training.

## Use cases
- generate photorealistic images from text prompts
- run text-to-image generation with Chinese and English prompts
- fine-tune a diffusion model with DreamBooth LoRA on custom subjects
- control image generation with pose, canny, or depth conditioning
- inpaint or edit regions of generated images
- preserve identity or faces in generated portraits with IP-Adapter

## When to choose
- you need high-quality photorealistic text-to-image generation with strong bilingual text rendering
- you want a diffusion model integrated with Hugging Face Diffusers
- you need ControlNet, IP-Adapter, or inpainting extensions for controllable image synthesis
- you want to fine-tune a text-to-image model on your own images with LoRA

## When to avoid
- you need a lightweight model that runs without a GPU
- you only need image classification or detection rather than image generation
- you require a fully self-hosted production image API out of the box rather than model code and weights

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, llm-training
- domain: artificial-intelligence, deep-learning, image-processing, computer-vision
- platform: python, cross-platform
- tags: text-to-image, diffusion-model, stable-diffusion, generative-ai, diffusers, controlnet, ip-adapter, inpainting, dreambooth, lora, photorealistic, huggingface, gpu, linux

## Member repositories
- Kwai-Kolors/Kolors (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:55.109653+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 2026-08-29T18:19:40.369083+00:00, confidence not recorded.
  - readme: https://github.com/Kwai-Kolors/Kolors (fetched 2026-08-28T04:08:55.109653+00:00, sha 70e95f88515a)
- Data as of 2026-08-30T08:39:29.467469+00:00.