PhotoGenius AI Models: Complete Technical Guide

PhotoGenius AI Models

PhotoGenius provides access to specialized AI models for different creative tasks. Each model has been selected for specific strengths in image generation, video creation, or content editing.

Model Categories

PhotoGenius organizes models by their primary function:

Image Generation: FLUX 1.1 Pro creates images from text descriptions Image Editing: Google Nano Banana modifies existing images Video Generation: Google Veo 2/3, WAN 2.1, Minimax O1, Kling 1.6 Pro Ad Creation: Nano Banana specializes in product photography

Image Generation Models

FLUX 1.1 Pro - Primary Image Generator

Technical Specifications:

• Resolution: Up to 1024x1024
• Generation time: 10-20 seconds at quality 90
• Aspect ratios: 1:1, 16:9, 9:16, 4:3
• Batch generation: 1-4 images simultaneously

Strengths:

• Photorealistic imagery
• Complex lighting scenarios
• Detailed textures and materials
• Consistent character features

Optimal Use Cases:

Product mockups, professional photography, 
architectural visualization, character portraits

FLUX Limitations:

• May struggle with text rendering in images
• Hand details can require multiple attempts
• Very long prompts (100+ words) may lose coherence

Google Nano Banana - Image Editor & Variations

Technical Specifications:

• Input: Existing images + text prompts
• Processing time: 5-15 seconds
• Supports masking for targeted edits
• Maintains original image resolution

Core Functions:

• Inpainting: Edit specific masked areas
• Variations: Generate similar images from input
• Style Transfer: Apply artistic styles to photos
• Object Replacement: Swap elements in scenes

When to Use Nano Banana:

Product color variations, background replacements,
object removal, style modifications, detail refinements

Nano Banana Best Practices:

• Keep mask areas reasonable (not too large)
• Describe only what should appear, not what to remove
• Use for iterative refinements after FLUX generation

Video Generation Models

PhotoGenius offers five specialized video models:

Google Veo 3 - Ultra Fast

Specifications:

• Generation speed: Fastest available
• Duration: 4-6 seconds
• Resolution: Up to 720p
• Best for: Quick previews, social media clips

Google Veo 2 - Realistic

Specifications:

• Focus: Photorealistic motion
• Duration: 6 seconds
• Aspect ratio: 16:9
• Best for: Realistic scenes, nature footage

WAN 2.1 - High Quality

Specifications:

• Frame count: 81 frames
• Resolution: 480p default
• Quality focus: Motion coherence
• Best for: Artistic animations

Minimax O1 - Cinematic

Specifications:

• Style: Film-quality output
• Prompt optimization: Built-in
• Supports starting images
• Best for: Movie-style content

Kling 1.6 Pro - Artistic

Specifications:

• Duration: 5 seconds
• Style: Artistic interpretation
• CFG scale: 0.5 default
• Best for: Creative, stylized videos

Model Selection Guide

For Images

Creating from scratch? → Use FLUX 1.1 Pro Editing existing image? → Use Google Nano Banana Need product photos? → Use Ad Generator (Nano Banana)

For Videos

Need it fast? → Google Veo 3 Want realism? → Google Veo 2 Cinematic look? → Minimax O1 Artistic style? → Kling 1.6 Pro Maximum quality? → WAN 2.1

Model-Specific Best Practices

FLUX 1.1 Pro Prompting

Structure that works:

[Style] + [Subject] + [Environment] + [Lighting] + [Quality modifiers]

Example:

"Photograph, businessman in modern office, 
natural window light, professional headshot, sharp focus"

FLUX responds well to:

• Photographic terminology ("DSLR", "35mm lens", "f/1.4")
• Lighting descriptions ("golden hour", "studio lighting")
• Composition terms ("rule of thirds", "centered")

Nano Banana Editing

For inpainting (masked edits):

1. Upload base image
2. Draw mask over area to change
3. Describe only what should appear in masked area
4. Keep prompts focused on the change

Example edit prompts:

"Red leather jacket" (when changing clothing)
"Sunset sky with clouds" (when changing background)
"Smiling expression" (when editing faces)

Video Model Prompting

Google Veo (2 & 3):

"Camera movement + Subject + Action + Style"
Example: "Drone shot of ocean waves crashing, cinematic"

Kling 1.6 Pro:

"Artistic style + Subject + Motion + Mood"
Example: "Watercolor style, flowers blooming, time-lapse, peaceful"

Quality Settings Guide

Image Generation (FLUX 1.1 Pro)

• 70-80: Draft quality, 5-10 seconds
• 85-90: Production quality, 10-15 seconds
• 95-100: Maximum quality, 15-25 seconds

Video Generation

• Google Veo 3: Optimized for speed over quality
• WAN 2.1: Highest quality, longest processing
• Others: Balanced quality/speed ratio

Technical Limitations

FLUX 1.1 Pro

• Text in images may be garbled
• Complex hand poses often need refinement
• Extremely detailed scenes may lose coherence

Nano Banana

• Large mask areas may produce inconsistent results
• Style transfer works best with clear artistic references
• Color changes maintain better consistency than structural changes

Video Models

• All models limited to short durations (5-8 seconds)
• Complex physics simulations may appear unrealistic
• Text/logos in videos often unstable

Workflow Recommendations

For Product Photography

1. Generate base with FLUX 1.1 Pro
2. Use Nano Banana for color variations
3. Apply Ad Generator for commercial layouts

For Social Media Content

1. Quick drafts with FLUX at quality 80
2. Refine winners at quality 95
3. Add motion with Google Veo 3 for stories

For Professional Projects

1. Generate multiple FLUX variations (batch of 4)
2. Select best, edit details with Nano Banana
3. Export at maximum quality settings

API and Integration Notes

All models accessible through PhotoGenius interface:

• Image Playground: FLUX 1.1 Pro, Nano Banana
• Video Playground: All video models
• Ad Generator: Nano Banana specialized for products

Generation costs vary by model complexity and quality settings. Monitor credit usage in account dashboard.

Technical specifications subject to updates. Check model selector for latest capabilities.