Single Image to Multiple Consistent Shots
Nano Banana
Reference Image
VLM
0
56
AI-Powered Multi-Prompt Image Generation Workflow
Overview
This advanced workflow leverages Vision Language Models (VLM) and state-of-the-art image editing APIs to generate multiple unique image variations from a single reference image through intelligent prompt extraction and iterative refinement.
Key Features
🎯 Intelligent Prompt Generation
Utilizes Google Gemini 2.5 Flash VLM to analyze reference images and generate 6 distinct, creative prompts
Each prompt is automatically crafted with varying camera angles, shot types, and subtle details while maintaining consistent core elements
Prompts are optimized to 50-80 words for maximum effectiveness
🔄 Automated Batch Processing
Seamlessly processes multiple prompts through regex-based extraction and indexing
Generates 6 unique image variations in a single workflow execution
Individual prompt selection capability for targeted regeneration
⚙️ Flexible Node Architecture
Modular design with reroute nodes for clean signal flow
Integrated math operations for prompt indexing (converts prompt numbers to zero-based indices)
String manipulation nodes for precise prompt parsing and formatting
🖼️ High-Quality Output
Powered by NanoBanana Edit API for professional-grade image generation
Supports JPEG output format with customizable parameters
Automatic image batching and organized file saving (image_1 through image_6)
Technical Specifications
Workflow Components:
VLM Integration: Google Gemini 2.5 Flash for intelligent prompt generation
Image Processing: NanoBanana Edit API for high-quality renders
Regex Operations: Advanced pattern matching for prompt extraction
Node Count: 67 optimally arranged nodes
Output: 6 simultaneous image generations with individual save nodes
Use Cases
Character concept art exploration with consistent styling
Product photography variations with different angles and contexts
Creative storytelling through sequential image generation
Marketing asset creation with diverse visual approaches
AI-assisted creative ideation and rapid prototyping
Workflow Advantages
Time-Efficient: Generate 6 professional variations in minutes instead of hours
Consistency: Maintains visual coherence across all generated images
Customizable: Easy prompt number selection for regenerating specific outputs
Production-Ready: Clean node organization for easy modification and extension
Scalable: Modular architecture allows for adding more prompt slots
This workflow represents a production-grade solution for creators, designers, and AI artists seeking to maximize creative output while maintaining artistic control and quality standards.
Read more
Nodes & Models
VLM_floyo
NanoBananaEdit_floyo
PrimitiveInt
WorkflowGraphics
LoadImage
Reroute
SomethingToString
PrimitiveStringMultiline
RegexReplace
RegexExtract
SaveImage
SimpleMath+
SimpleMath+
AI-Powered Multi-Prompt Image Generation Workflow
Overview
This advanced workflow leverages Vision Language Models (VLM) and state-of-the-art image editing APIs to generate multiple unique image variations from a single reference image through intelligent prompt extraction and iterative refinement.
Key Features
🎯 Intelligent Prompt Generation
Utilizes Google Gemini 2.5 Flash VLM to analyze reference images and generate 6 distinct, creative prompts
Each prompt is automatically crafted with varying camera angles, shot types, and subtle details while maintaining consistent core elements
Prompts are optimized to 50-80 words for maximum effectiveness
🔄 Automated Batch Processing
Seamlessly processes multiple prompts through regex-based extraction and indexing
Generates 6 unique image variations in a single workflow execution
Individual prompt selection capability for targeted regeneration
⚙️ Flexible Node Architecture
Modular design with reroute nodes for clean signal flow
Integrated math operations for prompt indexing (converts prompt numbers to zero-based indices)
String manipulation nodes for precise prompt parsing and formatting
🖼️ High-Quality Output
Powered by NanoBanana Edit API for professional-grade image generation
Supports JPEG output format with customizable parameters
Automatic image batching and organized file saving (image_1 through image_6)
Technical Specifications
Workflow Components:
VLM Integration: Google Gemini 2.5 Flash for intelligent prompt generation
Image Processing: NanoBanana Edit API for high-quality renders
Regex Operations: Advanced pattern matching for prompt extraction
Node Count: 67 optimally arranged nodes
Output: 6 simultaneous image generations with individual save nodes
Use Cases
Character concept art exploration with consistent styling
Product photography variations with different angles and contexts
Creative storytelling through sequential image generation
Marketing asset creation with diverse visual approaches
AI-assisted creative ideation and rapid prototyping
Workflow Advantages
Time-Efficient: Generate 6 professional variations in minutes instead of hours
Consistency: Maintains visual coherence across all generated images
Customizable: Easy prompt number selection for regenerating specific outputs
Production-Ready: Clean node organization for easy modification and extension
Scalable: Modular architecture allows for adding more prompt slots
This workflow represents a production-grade solution for creators, designers, and AI artists seeking to maximize creative output while maintaining artistic control and quality standards.
Read more
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(10)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(8)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(7)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(6)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(5)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(4)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(3)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(2)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(1)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)_1771484497818.png?width=1400&height=620&quality=80&resize=cover)

%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(10)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(8)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(7)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(6)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(5)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(4)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(3)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(2)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)%20(1)_1771484497818.png?width=104&height=104&quality=80&resize=cover)
%20(2400%20x%201080%20px)%20(2430%20x%201080%20px)_1771484497818.png?width=104&height=104&quality=80&resize=cover)