Google has launched two new AI models designed to make content creation faster and cheaper: Nano Banana 2 Lite for image generation and Gemini Omni Flash for video editing. Nano Banana 2 Lite, officially named Gemini 3.1 Flash-Lite Image, generates high-quality images in approximately 4 seconds while costing just over 3 cents per 1,000-pixel resolution image. This represents a significant speedup compared to its predecessor, Nano Banana 2, which takes about 20 seconds per image.

How Fast Is Nano Banana 2 Lite Compared to Previous Models?

The new model achieves a dramatic performance improvement across multiple dimensions. In real-world testing, Nano Banana 2 Lite generated images in an average of 3 seconds, compared to 19 seconds for Nano Banana 2. When evaluated on Arena.ai, a platform where humans compare AI-generated images, Nano Banana 2 Lite scored 1,251 on image generation quality, just 19 points below Nano Banana 2's score of 1,270 but 100 points higher than the original Nano Banana model's 1,151.

The pricing advantage is equally compelling. A single 1,000-pixel resolution image costs approximately 3 cents with Nano Banana 2 Lite, compared to 3.9 cents for Nano Banana 2 and 6.7 cents for the original model. This positions Nano Banana 2 Lite as a model that balances quality, speed, and affordability rather than pursuing maximum image quality at any cost.

Where Can Developers and Users Access These New Models?

Google is rolling out Nano Banana 2 Lite across multiple platforms and services. The model is immediately available through Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform. Google has announced plans to integrate it into consumer-facing services including Google Search's AI Mode, the Gemini app, NotebookLM, Google Photos, Google Ads, and other products.

NotebookLM, Google's AI-powered note-taking and research tool, is already leveraging Nano Banana 2 Lite to create Short Video Overviews. This feature transforms uploaded documents into approximately 60-second vertical videos designed for quick learning. The feature is currently rolling out to Google AI Pro and Google AI Ultra subscribers on both mobile and web platforms, with free NotebookLM users expected to gain access in the future.

What Is Gemini Omni Flash and How Does It Work?

Gemini Omni Flash is a multimodal video generation and editing model announced at Google I/O 2026. It enables users to create videos by combining text, images, and video inputs, with support for interactive editing through natural language instructions. The model can synchronize on-screen text and graphics with the movements of people and objects in videos, making it useful for correcting videos with written instructions and maintaining scene consistency by referencing images and text.

The model is available as a public preview through Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform, with integration into Gemini apps and Google Flow. Video generation costs 10 cents per second, which is competitive with Google's speed-focused video model Veo 3.1 Fast. Currently, Gemini Omni Flash can generate videos up to 10 seconds long, meaning maximum output costs would be approximately $1 per video, though Google plans to support longer videos in the future.

Steps to Get Started With Nano Banana 2 Lite and Gemini Omni Flash

• Access via Google AI Studio: Developers can immediately begin experimenting with Nano Banana 2 Lite and Gemini Omni Flash through Google's free AI Studio interface without requiring API credentials or setup.
• Integrate into Applications: Use the Gemini API to embed image and video generation capabilities into custom applications, with pricing of 25 cents per 1 million input tokens and $1.50 per 1 million output tokens for Nano Banana 2 Lite.
• Leverage NotebookLM for Document Conversion: Upload documents or notes to NotebookLM and select the Video tab in the Studio section to automatically generate Short Video Overviews for educational or explanatory purposes.
• Create Iterative Prototypes: Use Nano Banana 2 Lite's 4-second generation speed for rapid design reviews and prototyping where quick iteration is critical to the workflow.

What Safety Measures Does Google Include in These Models?

Google has implemented several safeguards to ensure responsible AI-generated content. Images created by Nano Banana 2 Lite are embedded with an invisible digital watermark called SynthID that identifies them as AI-generated. The company also filters, labels, and red-teams training data to suppress harmful outputs, demonstrating a commitment to ensuring identifiability as the volume of generated images increases with faster and cheaper generation.

Gemini Omni Flash does have some current limitations worth noting. The Gemini API does not yet support uploading audio references or extending scenes. While the API accepts video references up to 3 seconds long, the model may not process them correctly. Additionally, there is room for improvement in character consistency in videos involving scene changes and camera movements.

Why Does This Matter for Content Creators and Developers?

The combination of Nano Banana 2 Lite and Gemini Omni Flash addresses a key challenge in AI-assisted content creation: the tension between quality, speed, and cost. By achieving 4-second image generation at under 3 cents per image, Nano Banana 2 Lite makes high-volume image generation economically viable for use cases like design reviews, prototyping, and educational content. The integration into NotebookLM demonstrates how these models can enhance existing productivity tools, enabling users to transform text-based research into engaging visual content in minutes rather than hours.

For video creators, Gemini Omni Flash's ability to edit videos through natural language instructions and maintain consistency across scenes opens new workflows for rapid video iteration. At 10 cents per second, the pricing is accessible for small teams and individual creators experimenting with AI-assisted video production. The rollout across Google's ecosystem suggests these models will become increasingly integrated into everyday tools, making AI-assisted content creation a standard feature rather than a specialized capability.