Speed-optimized ERNIE Image variant with lower default step count for faster multilingual text-to-image generation.
Added Apr 13, 2026
Approx. Price
$0.010 per image
Model Type
text-to-image
Preview Examples
3
Generation controls available for this model.
Images Per Run
Up to 4
Resolution Options
11
512x512 (Square), 1024x1024 (Square HD), 768x1024 (Portrait (3:4)) +8 more
Tunable Settings
9
Enable Safety Checker
Default
No
Guidance Scale
Default
1
How strongly to follow the prompt (1-20).
Inference Steps
Default
8
Number of denoising steps (1-20).
Negative Prompt
Default
N/A
Describe what to avoid in the image.
Number of Images
Default
1
Output Format
Default
jpeg
Options (2)
JPEG, PNG
Image output format.
Prompt Expansion
Human preference benchmarks sourced from Artificial Analysis.
No Artificial Analysis benchmark data is available yet for this model.
Artificial Analysis API
Minimalist product photo of matte-black wireless earbuds on a floating glass pedestal, soft studio lighting, high contrast shadows, clean white background
512x512, default turbo settings
中国古风插画：月光下的江南水乡，小桥、乌篷船、灯笼倒影，细腻笔触，电影级光影
512x512, default turbo settings
Street-style fashion photo of a skater jumping over a puddle at blue hour, neon reflections, dynamic motion blur, cinematic framing
512x512, default turbo settings
Default
Yes
Enhance the prompt automatically for richer outputs.
Resolution
Default
1024x1024
Options (11)
512x512 (Square), 1024x1024 (Square HD), 768x1024 (Portrait (3:4)), 576x1024 (Portrait (9:16)) +7 more
Seed
Default
N/A
Random seed for reproducible outputs.
