Google Unveils Nano Banana 2: Gemini 3.1 Flash Image Takes Center Stage

Google has officially released Nano Banana 2, an updated AI image generator also known as Gemini 3.1 Flash Image. This new version succeeds Gemini 2.5 Flash Image (Nano Banana) and Gemini 3 Pro Image (Nano Banana Pro), introducing several significant enhancements focused on improving image generation quality, user control, and expanding capabilities previously exclusive to Pro models.

Nano Banana 2 is designed to offer more precise instruction following and improved visual output, integrating advanced features to enhance subject consistency and text rendering within generated images.

Google specifies that Nano Banana 2 is optimized for rapid image generation, aiming to provide a quicker and more efficient user experience.

Overview of Nano Banana 2

This latest iteration of Google's AI image generation technology brings substantial improvements. It is engineered to deliver images that are not only visually superior but also adhere more closely to user prompts. The focus remains on enhancing overall quality while making advanced features more accessible.

Key Features and Enhancements

The model includes several improvements aimed at refining the image generation process:

Nano Banana 2 can maintain the appearance of up to five characters and the fidelity of up to 14 objects within a single workflow. This robust feature significantly supports storyboarding and narrative creation, allowing users to develop sequences without altering initial input appearances.

The model now exhibits higher adherence to user instructions, effectively capturing nuances in complex prompts. This leads to images that more accurately reflect the intended idea, providing users with greater creative control and predictability.

Generated images feature improved lighting, richer textures, and sharper details. Outputs are available in various aspect ratios and resolutions, ranging from 512 pixels to a high-definition 4K. Supported aspect ratios include:

• 1:1 (square)
• 16:9 (widescreen)
• 9:16 (vertical)

This versatility accommodates diverse applications, from vertical social media posts to widescreen backdrops, ensuring flexibility for all user needs.

Integrated Pro Features

Nano Banana 2 incorporates two powerful features previously found exclusively in the Nano Banana Pro model:

This capability allows the generator to utilize real-time information and images from the web. It significantly aids in the creation of infographics, diagrams, data visualizations, and marketing mockups by ensuring more precise rendering of specific subjects and concepts.

Text within generated images is designed to have fewer typographical errors and enhanced clarity. The model also supports text translation within images, a valuable tool for global content creation and accessibility.

Availability and Access

Nano Banana 2 is accessible to all users through multiple Google platforms, with no subscription required for access. Paid users, however, may experience broader access or additional benefits.

The model is available via:

• Gemini app
• AI Mode
• Google Lens
• AI Studio (in preview)
• Google Cloud
• Flow (as the default image generation model)
• Google Ads
• Gemini API (in preview)
• Google Antigravity (in preview)
• Vertex AI (in preview)
• Gemini CLI (in preview)

Changes for Existing Users and Model Distinction

In the Gemini app, Nano Banana 2 replaces Nano Banana Pro across all models (Fast, Thinking, and Pro). Google AI Pro and Ultra subscribers will retain access to Nano Banana Pro for specialized tasks. To utilize Nano Banana Pro, users must first generate an output with Nano Banana 2 and then select "Redo with Pro" from the three-dot menu regeneration option.

Google clearly distinguishes between the two models: Nano Banana 2 is optimized for "rapid generation," while Nano Banana Pro is intended for "high-fidelity tasks requiring maximum factual accuracy." This differentiation allows users to choose the most appropriate tool for their specific creative needs.

Additional Features and Safety Measures

A new templates feature has been introduced within the Gemini app to assist users with selecting various image styles, streamlining the creative process.

Regarding safety, Google is implementing SynthID watermarks in conjunction with C2PA Content Credentials. This combination aims to provide users with a comprehensive view of AI usage, indicating whether AI was used and how. C2PA verification functionality is scheduled for integration into the Gemini app, further enhancing transparency and trust.