[Report] Thorough Comparison of Gemini and GPT in Image Generation and Character Consistency
Combeez Inc. has released the results of a comparative study on image generation and character consistency using the leading generative AI models "Gemini 3.5 Flash" and "GPT-5.5 Instant."
📋 Article Processing Timeline
- 📰 Published: May 26, 2026 at 22:00
- 🔍 Collected: May 26, 2026 at 13:31
- 🤖 AI Analyzed: May 26, 2026 at 13:39 (8 min after Collected)
## Survey Overview
Combeez Inc. conducted a comparative study on image generation and character consistency using major generative AI models: Google's "Gemini 3.5 Flash" and OpenAI's "GPT-5.5 Instant."
In this study, the company utilized images of its original mascot, "Combee-chan," and analyzed the output results and specific strengths and weaknesses of each AI based on two patterns: a "simple prompt" and a "complex prompt."
## Survey 1: Comparison via Broad Instructions
For the prompt: "Generate an illustration of the attached character flying through the city."
Both AIs successfully grasped the intention from broad instructions and produced illustrations of the character flying through the city without major issues. Both possess high practical utility for simple image generation tasks where the AI is left with some creative freedom.
## Survey 2: Comparison via Complex Instructions
We conducted a survey using highly challenging prompts that specified situations, touch, and composition in detail. Distinct differences in features and capabilities became apparent with detailed instructions.
Regarding overall quality and character consistency, GPT demonstrated superior performance. Gemini showed a slight tendency to struggle with reflecting all detailed instructions in the illustration (e.g., character having limbs when it shouldn't).
Additionally, distinct individualities emerged in drawing styles and texture. While Gemini produced warm illustrations with clear line art similar to a picture book, GPT excelled in expressing realistic textures, faithfully reproducing the characteristics of watercolor paintings.
## Conclusion
As of now, GPT demonstrates superior ability in detailed character consistency and complex composition specifications. However, AI is evolving daily, and there are high expectations for further improvements in functionality.
## Survey Summary
- Target: Major Generative AI (Gemini 3.5 Flash / GPT-5.5 Instant)
- Period: April 17, 2026 – May 16, 2026
- Purpose: Investigation of output features based on prompt complexity.
Combeez Inc. conducted a comparative study on image generation and character consistency using major generative AI models: Google's "Gemini 3.5 Flash" and OpenAI's "GPT-5.5 Instant."
In this study, the company utilized images of its original mascot, "Combee-chan," and analyzed the output results and specific strengths and weaknesses of each AI based on two patterns: a "simple prompt" and a "complex prompt."
## Survey 1: Comparison via Broad Instructions
For the prompt: "Generate an illustration of the attached character flying through the city."
Both AIs successfully grasped the intention from broad instructions and produced illustrations of the character flying through the city without major issues. Both possess high practical utility for simple image generation tasks where the AI is left with some creative freedom.
## Survey 2: Comparison via Complex Instructions
We conducted a survey using highly challenging prompts that specified situations, touch, and composition in detail. Distinct differences in features and capabilities became apparent with detailed instructions.
Regarding overall quality and character consistency, GPT demonstrated superior performance. Gemini showed a slight tendency to struggle with reflecting all detailed instructions in the illustration (e.g., character having limbs when it shouldn't).
Additionally, distinct individualities emerged in drawing styles and texture. While Gemini produced warm illustrations with clear line art similar to a picture book, GPT excelled in expressing realistic textures, faithfully reproducing the characteristics of watercolor paintings.
## Conclusion
As of now, GPT demonstrates superior ability in detailed character consistency and complex composition specifications. However, AI is evolving daily, and there are high expectations for further improvements in functionality.
## Survey Summary
- Target: Major Generative AI (Gemini 3.5 Flash / GPT-5.5 Instant)
- Period: April 17, 2026 – May 16, 2026
- Purpose: Investigation of output features based on prompt complexity.
FAQ
How was the comparison study between Gemini and GPT conducted?
The company's character 'Konbi-chan' was used to test two scenarios: simple prompts and complex prompts with detailed specifications.
What were the differences in the results of image generation between the two AIs?
Both showed high practicality with simple instructions, but GPT demonstrated higher character reproducibility with complex detailed specifications.
What were the differences in style and texture tendencies?
Gemini tends to generate illustrations with solid lines and a warm feel, while GPT excels at reproducing realistic textures and the characteristics of watercolor paintings in detail.
What AI models were used in the study?
Google's 'Gemini 3.5 Flash' and OpenAI's 'GPT-5.5 Instant'.
What was the purpose of this study?
To investigate how the complexity of instructions (prompts) to AI affects the output results and the strengths and weaknesses of AI.