ziqipang/RandAR

[CVPR 2025 (Oral)] Open implementation of "RandAR"

35
/ 100
Emerging

RandAR is a tool for advanced image generation, offering highly flexible control over how images are created. It takes basic visual cues or prompts and produces diverse images, enabling tasks like filling in missing parts of an image, extending its borders, or generating new visual content from scratch. This is ideal for researchers and practitioners in computer vision who need highly adaptable image synthesis capabilities.

207 stars. No commits in the last 6 months.

Use this if you need a flexible image generation model that can handle various visual tasks like inpainting, outpainting, and creating high-resolution images from basic inputs, without needing specific fine-tuning for each task.

Not ideal if you are looking for a simple, off-the-shelf image generator for casual use without delving into advanced model capabilities or research.

image-generation computer-vision-research generative-ai image-synthesis visual-content-creation
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

207

Forks

6

Language

Python

License

MIT

Last pushed

Jul 14, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/ziqipang/RandAR"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.