Google has launched Gemini 2.5 Flash Image, also known as “nano-banana,” enhancing its AI-driven image generation and editing capabilities. This model targets developers seeking greater creative control and improved image quality, addressing limitations seen in the prior Gemini 2.0 Flash. Key features include multi-image fusion for realistic compositions, character consistency across varied settings, and targeted transformations via natural language prompts for precise edits. Additionally, the model integrates world knowledge for educational assistance and adheres to templates for uniform asset generation. Priced at $30 per million output tokens (approximately $0.039 per image), Gemini 2.5 also includes upgraded developer tools for speedy app deployment and project management via GitHub. To enhance adoption, partnerships with OpenRouter.ai and fal.ai expand accessibility to over three million developers. Incorporating SynthID for invisible watermarking labels all AI-generated images. Continuous improvements in long-form text and character consistency are underway, with a stable release anticipated soon.
Source link
Unleashing the Power of Google’s Gemini 2.5 Flash Image: Mastering Background Blur and Multi-Image Fusion
Share
Read more