Google launches Nano Banana 2 and changes image generation rules

Google Launches Nano Banana 2 and Updates Image Generation Guidelines

Google DeepMind has introduced a new image generation model, Nano Banana 2, which combines the advanced features of Nano Banana Pro with the high speed of Gemini Flash. This model allows users to create high-quality images with quick editing and iteration on Google platforms such as the Gemini app and Google Search. Additionally, Google is enhancing its SynthID technology with C2PA Content Credentials for more accurate identification of AI-generated content.

Key Points

• Nano Banana 2 combines speed and advanced features of Google's best image models.
• The model provides fast image generation, simplifying editing.
• Users gain enhanced creative control, ensuring object consistency and precise execution of instructions.
• The model is integrated into Google platforms, including Gemini, Search, and Ads.
• Google is advancing AI image identification with SynthID and C2PA Content Credentials.

Introduction

Google's new AI image generation model, Nano Banana 2, stands out with its high speed and integration of the best aspects of previous models. It enables faster creation of high-quality images with better control and is available in Google applications such as Gemini and Google Ads.

Last August, the Gemini image model, Nano Banana, became a viral sensation, redefining image generation. In November, Nano Banana Pro was released, offering sophisticated intelligence and studio-quality control. Now, Google combines these features in Nano Banana 2 (Gemini 3.1 Flash Image), providing the advanced knowledge, quality, and logic of Nano Banana Pro at unprecedented speed.

Intelligence and Visual Quality at Flash Speed

Nano Banana 2 incorporates the high-speed intelligence of Gemini Flash for visual generation, allowing for quick editing and iteration. This makes Pro features more accessible, including:

• Advanced world knowledge: Utilizing Gemini's real-world knowledge base, the model uses information and images from web searches for accurate object depiction. It also aids in creating infographics, converting notes into diagrams, and generating data visualizations.
• Precise text rendering and translation: Enables accurate and legible text generation for marketing materials or postcards, supporting text translation and localization within images for global sharing.

Enhanced Creative Control

Nano Banana 2 bridges the gap between speed and visual accuracy, providing photorealistic images. Key improvements include:

• Object consistency: Ensures character likeness and object accuracy in workflows, aiding in storyboarding and narrative building.
• Precise instruction execution: Improved adherence to complex requests, capturing the nuances of user ideas.
• Production readiness: Control over aspect ratios and resolutions from 512px to 4K, ensuring visual clarity for various applications.
• Updated visual accuracy: Provides vibrant lighting, richer textures, and sharper details at Flash speed.

Try Nano Banana 2 Today

Now available in Google products, Nano Banana 2 is suitable for any workflow, from rapid generation and precise instruction execution to image search integration. It is available in:

• Gemini App: Replaces Nano Banana Pro in Fast, Thinking, and Pro models. Google AI Pro and Ultra subscribers can still access Nano Banana Pro for specific tasks.
• Search: Available in AI Mode and Lens via the Google app and browsers, now in 141 countries and eight languages.
• AI Studio + API: Available in preview, with pricing and access information online.
• Google Cloud: Available in preview via the Gemini API in Vertex AI.
• Flow: Now the default image generation model, free for all Flow users.
• Google Ads: Available to enhance campaign creation with suggestions.

Try Nano Banana 2 in the Gemini app using the new template feature.

Reliable Provenance Verification: Tagging and Verification

As generative media evolves, our tools for identification and understanding must also advance. By integrating SynthID technology with C2PA Content Credentials, users gain a comprehensive view of AI usage. Since its launch, the SynthID verification feature has been used over 20 million times, helping users identify AI-generated content. C2PA verification will soon be added to the Gemini app.