Google has announced the update of its Gemini chatbot with a new AI image model, called Gemini 2.5 Flash Image, which allows users to fine-tune photo edits and preserve consistency in images. The new tool is designed to improve visual quality and follow instructions accurately, setting it apart from rival tools like ChatGPT and xAI’s Grok.
Gemini’s new AI image model is state-of-the-art on platforms such as LMArena and has already drawn attention for its impressive performance. Google claims that the model is behind the native image capability within Gemini 2.5 Flash, showcasing its capabilities in a crowdsourced evaluation platform.
The update aims to catch up with OpenAI’s popular image tools and draw users away from ChatGPT. To achieve this, Google has designed the AI image model with consumer use cases in mind, such as visualizing home and garden projects. The model also features better “world knowledge” and can combine multiple references in a single prompt.
However, while Gemini’s new AI image generator makes it easier for users to create realistic images, the company has implemented safeguards to prevent non-consensual intimate imagery. Google applies visual watermarks to AI-generated images and identifiers in metadata to address the rise of deepfake imagery.
The update is set to roll out on Tuesday to all users in the Gemini app and developers via the Gemini API, Google AI Studio, and Vertex AI platforms. With this new tool, Google hopes to close its user gap with OpenAI and compete in the critical battleground for Big Tech’s AI image models.
Source: https://techcrunch.com/2025/08/26/google-geminis-ai-image-model-gets-a-bananas-upgrade