Image Generation
Create images from text prompts using multiple AI providers. Compare results across DALL-E, Midjourney, Stability AI, and more.
Available Models
OpenAI
- DALL-E 3 - High quality, photorealistic
- DALL-E 2 - Fast, cost-effective
Stability AI
- Stable Diffusion XL - Versatile, controllable
Other Providers
Multiple image models available through our platform
How to Use
📸 Image Generation Interface
Image generation interface showing prompt input and model selection
📸 SCREENSHOT NEEDED: image-gen-interface.png
What to capture: Image generation interface (modal or page)
Show: Prompt text field, model selector dropdown, size options, Generate button
Tips: Open image generation, enter a prompt but don't generate yet
What to capture: Image generation interface (modal or page)
Show: Prompt text field, model selector dropdown, size options, Generate button
Tips: Open image generation, enter a prompt but don't generate yet
📸 Model Comparison
Same prompt generated with DALL-E 3 (left) and Stable Diffusion XL (right)
📸 SCREENSHOT NEEDED: image-gen-comparison.png
What to capture: Two generated images side-by-side
Show: DALL-E 3 result on left, Stable Diffusion XL result on right
Prompt: Use "Modern minimalist office with large windows, natural light"
Tips: Generate with both models, capture both results together
What to capture: Two generated images side-by-side
Show: DALL-E 3 result on left, Stable Diffusion XL result on right
Prompt: Use "Modern minimalist office with large windows, natural light"
Tips: Generate with both models, capture both results together
From Lab Chat
- Start a conversation
- Click "Generate Image" button
- Enter your prompt or use AI response as prompt
- Select image model
- Choose size and options
- Click "Generate"
Direct Image Generation
- Access image generation from navigation
- Enter detailed prompt
- Select model and size
- Generate and download
Writing Good Prompts
Be Specific
❌ "A dog"
✅ "A golden retriever puppy playing in autumn leaves, soft lighting, photograph style"
Include Details
- Subject: What's the main focus
- Style: Photorealistic, artistic, cartoon, etc.
- Lighting: Natural, dramatic, soft
- Composition: Close-up, wide shot, portrait
- Mood: Calm, energetic, mysterious
Example Prompts
"Modern minimalist office with large windows, natural light, plants, clean lines"
"Futuristic cityscape at sunset, flying cars, neon lights, cyberpunk style"
"Close-up portrait of elderly woman, natural lighting, warm tones, photograph"
Comparing Models
Different models excel at different styles:
DALL-E 3
- Best for: Photorealism, complex scenes
- Speed: Fast (~10-15 seconds)
- Cost: Moderate
Stable Diffusion XL
- Best for: Artistic styles, flexibility
- Speed: Fast (~8-12 seconds)
- Cost: Low
Costs
Image generation pricing varies by model:
- DALL-E 3: ~$0.04-0.08 per image
- DALL-E 2: ~$0.02 per image
- Stable Diffusion XL: ~$0.01-0.02 per image
Check current pricing in-app before generating.
Tips
- Start with detailed prompts
- Try multiple models for comparison
- Save good prompts for reuse
- Download images immediately
- Experiment with different styles
Limitations
- Images are subject to content policies
- Some models have size restrictions
- Generation can take 10-30 seconds
- NSFW content is blocked
Next Steps
- Lab Chat - Generate images in conversations
- Cost Forecasting - Track image generation costs
- Snapshots - Save conversations with images
Previous: ← Snapshots | Next: Cost Forecasting →