ERNIE Image is an open-weight text-to-image model built on an 8B Diffusion Transformer, optimized for text accuracy and structured layouts.

Overview

ERNIE Image is Baidu’s open-weight text-to-image model, designed to generate images with accurate in-image text, structured layouts, and complex multi-object compositions.

Key Features

• Generate clean, readable text inside images
• Create structured layouts like posters and comics
• Handle complex prompts without losing detail
• Support both realistic and stylized image generation
• Run locally on a single consumer GPU
• Improve results automatically with Prompt Enhancer

How It Works

ERNIE Image works by using a Diffusion Transformer backbone to generate images from text prompts. The model can be run locally on a single consumer GPU, and it includes a built-in Prompt Enhancer that expands short inputs into richer, structured prompts.

Benefits

• Accurate text rendering
• Structured layouts
• Complex prompt handling
• Fast generation
• Commercial usage rights

Use Cases

• Generating product images with accurate text
• Creating posters and comics with structured layouts
• Handling complex prompts for custom images

Integrations

• ComfyUI

Pricing Overview

ERNIE Image offers a simple pricing plan with credits that never expire. The pricing includes a Starter plan for $9.9, a Pro plan for $29.9, and a Scale plan for $49.9.