baidu / ERNIE-Image
ERNIE-Image
A high-quality text-to-image model developed by Baidu that supports English, Chinese, and Japanese prompts, with automatic prompt expansion built in.
# ERNIE-Image
A high-quality text-to-image model developed by Baidu that supports English, Chinese, and Japanese prompts, with automatic prompt expansion built in.
## Highlights
- High quality: Compact 8B model that rivals much larger open-weight models in visual quality.
- Strong text rendering: Renders clean, accurate typography in posters, infographics, and text-heavy visuals.
- Great instruction following: Handles complex prompts with multiple objects, detailed relationships, and precise requirements.
- Structured visuals: Excellent for posters, comics, storyboards, and multi-panel layouts.
- Wide style coverage: From realistic photography to design-oriented and cinematic looks.
## When to Use
- When top visual quality and prompt fidelity matter most.
- For posters, covers, ads, and other content where both aesthetics and accuracy are critical.
- When you prioritize final quality over generation speed.
API quick start
Run ERNIE-Image with a single API call.
{
"prompt": "This is a photograph depicting an urban s…",
"negativePrompt": "...",
"steps": 50,
"scale": 4.0
}