On August 26, 2025, Google officially released Gemini 2.5 Flash Image, a new image generation and editing model designed to give developers enhanced creative control and efficiency when building applications with AI-generated imagery. The model is available through the Gemini API, Google AI Studio, and Vertex AI.
What are Gemini 2.5 Flash Image’s key features
- Character and object consistency: Gemini 2.5 Flash Image can maintain the appearance of a character or object across multiple prompts and edits, supporting use cases like storytelling, product showcases, and brand asset generation.
- Targeted image editing: The model allows for precise, local edits with natural language prompts. Tasks such as blurring backgrounds, removing unwanted elements, recoloring images, or altering poses are now possible directly in the workflow.
- Multi-image blending: Developers can blend several input images into a single output, useful for placing objects into new environments, restyling scenes, or merging content in photorealistic ways.
- Integration with Gemini’s world knowledge: The model leverages Google’s broader Gemini AI to interpret and edit images with a deeper understanding of real-world concepts, supporting advanced applications like education, diagram interpretation, and semantic edits.
- Template-based design: The model adheres to visual templates, making it suitable for batch-generating uniform assets, such as employee badges, real estate cards, or product mockups, from a single design.
- Developer accessibility: Gemini 2.5 Flash Image is available for immediate use in preview via the Gemini API and Google AI Studio, with full stability expected in the coming weeks. Sample apps and templates are provided for rapid experimentation in Google AI Studio.
Pricing and availability
- Pricing: The model is priced at $30.00 per 1 million output tokens, with each image typically costing $0.039.
- Watermarking: All images created or edited with Gemini 2.5 Flash Image include an invisible SynthID watermark, marking them as AI-generated.
- Partnerships: Google has partnered with OpenRouter.ai and fal.ai to make the model accessible to a broader developer community.
Gemini 2.5 Flash Image is positioned as a tool for developers seeking more advanced, controllable, and integrated solutions for AI-generated imagery, with ongoing improvements planned for long-form text rendering and factual accuracy in images.
For more information visit the official announcement on Gemini 2.5 Flash Image on the Google Developer Blog
