ERNIE — China’s AI Image Generator Is Completely FREE
AI image generators are all the rage right now. While Dall-E2, MidJourney, and Stable Diffusion are currently center stage, there’s another AI model that’s been making the rounds on the internet.
ERNIE-ViLG is an open-source AI image generator developed by the Chinese tech giant Baidu.
What is ERNIE-ViLG?
ERNIE stands for Enhanced Representation through Knowledge Integration — Vision Language Generation. It is a unified generative pre-training framework for a bidirectional image-text generation with transformer model.
To be clear, ERNIE is not a Stable Diffusion variation. Although both models are open-source.
According to this paper, ERNIE-ViLG 2.0 is an improved text-to-image diffusion model with a knowledge-enhanced mixture of denoising experts, to incorporate extra knowledge about the visual scene and decouple the denoising capabilities in different steps.