modelscope/facechain

Health v2 · maintenance only

Adoption not part of the score

9508 stars · 873 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

FaceChain is a deep-learning toolchain from ModelScope for generating identity-preserved personal portraits (digital twins) from a single photo in seconds. It supports text-to-image and inpainting pipelines, integrates with ControlNet and LoRAs, and can be used via Python scripts, a Gradio interface, or SD WebUI.

Use cases

• generate personal portrait photos in different styles from one selfie
• create a digital twin avatar of myself
• identity-preserving portrait generation with stable diffusion
• run face-adapted text-to-image with ControlNet and LoRA support
• train custom portrait style LoRAs
• generate AI avatars for profile pictures

When to choose

• you need fast, identity-preserving portrait generation from a single photo
• you want compatibility with existing Stable Diffusion LoRAs and ControlNets
• you want both Python API and Gradio/WebUI interfaces

When to avoid

• you need general-purpose image editing unrelated to human portraits
• you lack a GPU or cannot run diffusion models locally
• you need a fully hosted no-setup solution without any model downloads

Facets

library · maturity active

deep-learning image-processing machine-learning artificial-intelligence image-processing deep-learning python cross-platform portrait-generation digital-twin text-to-image stable-diffusion controlnet lora gradio face-adapter gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="modelscope/facechain")