Kandinsky 3: Text-to-Image Synthesis for Multifunctional Generative Framework

TL;DR

Kandinsky 3 is a versatile, high-quality text-to-image diffusion model that supports multiple image generation tasks and is optimized for efficiency, with publicly available code and user-friendly demo.

Contribution

Introducing Kandinsky 3, a multifunctional T2I model with a simple architecture, extended capabilities, and a faster distilled version without quality loss.

Findings

High-quality, photorealistic image generation

Supports diverse tasks like inpainting, fusion, and video synthesis

Faster inference with maintained image quality

Abstract

Text-to-image (T2I) diffusion models are popular for introducing image manipulation methods, such as editing, image fusion, inpainting, etc. At the same time, image-to-video (I2V) and text-to-video (T2V) models are also built on top of T2I models. We present Kandinsky 3, a novel T2I model based on latent diffusion, achieving a high level of quality and photorealism. The key feature of the new architecture is the simplicity and efficiency of its adaptation for many types of generation tasks. We extend the base T2I model for various applications and create a multifunctional generation system that includes text-guided inpainting/outpainting, image fusion, text-image fusion, image variations generation, I2V and T2V generation. We also present a distilled version of the T2I model, evaluating inference in 4 steps of the reverse process without reducing image quality and 3 times faster than the base model. We deployed a user-friendly demo system in which all the features can be tested in the public domain. Additionally, we released the source code and checkpoints for the Kandinsky 3 and extended models. Human evaluations show that Kandinsky 3 demonstrates one of the highest quality scores among open source generation systems.