Nano Banana 2: Google’s latest AI image generation model

Google DeepMind is launching Nano Banana 2, a new image model that combines the advanced features of Nano Banana Pro with the speed of Gemini Flash. Users can now generate high-quality images with faster editing and iteration across Google products like the Gemini app and Google Search. Google is also improving its SynthID technology with C2PA Content Credentials to identify AI-generated content.

Intelligence and visual quality at Flash speed

Nano Banana 2 offers Gemini Flash’s high-speed intelligence for visual generation, enabling rapid edits. It brings Pro features to a wider audience:

• Advanced world knowledge: The model uses Gemini’s real-world knowledge and real-time web search data to accurately render subjects, create infographics, turn notes into diagrams, and generate data visualizations.
• Precision text rendering and translation: Nano Banana 2 allows for accurate, legible text in images for marketing mockups or greeting cards, and can even translate and localize text within an image for global sharing.

Enhanced creative control

Nano Banana 2 improves on the original Nano Banana by offering:

• Subject consistency: Maintain character resemblance for up to five characters and the fidelity of up to 14 objects in a single workflow, enabling storyboarding and narrative building without altering appearance.
• Precise instruction following: The model adheres more closely to complex requests, capturing nuances for accurate image generation.
• Production-ready specs: Create assets with control over various aspect ratios and resolutions from 512px to 4K, ensuring sharp visuals for any format.
• Visual fidelity upgrade: Nano Banana 2 delivers vibrant lighting, richer textures, and sharper details at high speeds.

Try Nano Banana 2 today

Nano Banana 2 is available in Google products, including:

• Gemini app: Replaces Nano Banana Pro across Fast, Thinking, and Pro models. Gemini Pro and Ultra subscribers retain access to Nano Banana Pro for specialized tasks by regenerating images.
• Search: Available in AI Mode and Lens via the Google app and browsers. Check availability and supported languages here.
• AI Studio + API: Available in preview in AI Studio and Gemini API. Pricing details are available here. Also accessible via Google Antigravity.
• Google Cloud: Available in preview via the Gemini API in Vertex AI.
• Flow: The new default image generation model in Flow, available for all users at no cost.
• Google Ads: Powers suggestions for creating campaigns.

Robust provenance: marking and verification

Google is enhancing its provenance approach by integrating SynthID technology with C2PA Content Credentials for a more comprehensive understanding of AI-generated content. The SynthID verification feature in the Gemini app has been used over 20 million times since its November launch, helping users identify Google AI-generated media. C2PA verification will soon be available in the Gemini app.

---

Posted in: AI, Gemini models

Author: Naina Raisinghani, Product Manager, Google DeepMind

Published: Feb 26, 2026

Related stories:

• Grow with Google: Google and the Massachusetts AI Hub are launching a new AI training initiative for the Commonwealth.
• Translate: Get more context and understand translations more deeply with new AI-powered updates in Translate.
• AI: Build with Nano Banana 2, our best image generation and editing model.
• Android: A more intelligent Android on Samsung Galaxy S26.
• Search: See the whole picture and find the look with Circle to Search.
• Gemini models: Gemini 3.1 Pro: A smarter model for your most complex tasks.

---

Image Captions:

• A flat lay infographic depicting the water cycle.
• Triptych infographic comparing cloud types.
• Museum Clos Lucé in Synthetic Cubism style.
• Localized “Native Wildlife” sign.
• Joyful characters and items at a farm.
• Fluffy friends building a treehouse.
• Misty panoramic aerial shot of a verdant valley.
• Highly stylized pop-art fashion portrait in different aspect ratios.

Link bài viết gốc

• Tags:
• Ai
• February 2026
• Blog.google