Alchemist: Turning Public Text-to-Image Data into Generative Gold

Pre-training equips text-to-image (T2I) models with broad world knowledge, but this alone is often…


Yandex

OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data

GPT-4o-level stylization consistency using only 2,600 pairs + 500 GPU hours!


Show Lab

New Tools for Brand Style Consistency and Control

Built to help creative teams explore, define, and apply a distinct brand style across every…


Recraft

Step1X-Edit: A Practical Framework for General Image Editing

We release a state-of-the-art image editing model, Step1X-Edit, which can provide comparable…


StepFun

DreamO: A Unified Framework for Image Customization

We propose DreamO, a unified image customization framework, which covers ID, IP, Tryon, and style…


ByteDance

JSON Style Templates for ChatGPT

Launching — JSON Visuals for ChatGPT . 50+ unique aesthetic codes, with attribute randomiser to…


Rahul Chakraborty

DreamID achieves high-fidelity face swapping

DreamID: A Fast and High-Fidelity diffusion-based Face Swapping via Triplet ID Group Learning


ByteDance

SliderSpace: Decomposing the Visual Capabilities of Diffusion Models

Instead of relying on artist names or style descriptions, SliderSpace automatically can map out how…


Rohit Gandikota...

Privacy Preference Center