Google launches Gemini 2.5 Flash Image model in general availability

Google has officially launched Gemini 2.5 Flash Image, a production-ready image generation and editing model now available to developers and enterprises. This release opens access to a broader audience through the Gemini API, Google AI Studio, and Vertex AI, supporting both individual creators and organizations looking for scalable visual solutions. The model is now generally available and not limited to a select group, making it accessible for global use wherever these Google platforms operate.

Gemini 2.5 Flash Image introduces technical advancements such as support for 10 aspect ratios, including landscape, portrait, square, and flexible formats, allowing for tailored content creation across different media types. Users can blend multiple images, maintain character consistency for narrative work, and perform precise, natural language edits, all while leveraging Gemini’s knowledge base. Compared to previous models, Gemini 2.5 Flash Image is distinguished by its ability to render characters from any camera angle without sacrificing world knowledge or pose fidelity, addressing challenges faced by platforms like Cartwheel. The system operates at low latency, typically under 10 seconds, enabling real-time applications as demonstrated by Volley in their AI-powered game sessions.

Developers and companies worldwide are already integrating Gemini 2.5 Flash Image into creative workflows, educational tools, and live interactive experiences. Google’s new pricing structure sets the cost at $0.039 per image and $30 per million output tokens, maintaining competitive rates for enterprise adoption.

Source