Notices
Usage Limitations
Language Support
For optimal performance, please use the following languages:
- English
- Spanish (Español)
- Japanese (日本語)
- Chinese (中文)
- Hindi (हिन्दी)
Input Restrictions
- No audio or video input support: Image generation models only accept text prompts
- Uncertain output quantity: The model may not always generate the exact number of images as explicitly requested by users
- Optimal output limit: This model performs best when outputting a maximum of 3 images
Text Generation Recommendations
When generating text for images, it's best to generate the text first, then request an image containing that text for better results.
Regional Restrictions
Currently, the following regions do not support uploading children's photos:
- European Economic Area (EEA)
- Switzerland (CH)
- United Kingdom (UK)
Watermark Identification
All generated images contain SynthID watermarks to identify AI-generated content.
Usage Recommendations
Prompt Optimization
- Use specific, detailed descriptions
- Avoid vague or overly broad vocabulary
- Include key information about style, atmosphere, and colors
- Specify camera angles and composition requirements
Quality Control
- Try multiple times to achieve optimal results
- Use iterative approaches for gradual refinement
- Reference professional photography and art terminology
- Clearly specify image purpose and context
Content Compliance
- Follow platform terms of service
- Avoid generating inappropriate content
- Respect copyright and intellectual property
- Be mindful of cultural sensitivity issues
Technical Information
Model Characteristics
- Based on deep learning image generation technology
- Supports various artistic and photographic styles
- Has contextual understanding capabilities
- Allows conversational adjustments and optimization
Performance Optimization
- Recommend using supported languages for best results
- Avoid overly complex multi-element scene descriptions
- Reasonably control prompt length
- Utilize templates and best practices