Skip to content
4ALL API Docs
Main Site News Console

Get Started Building with Nano Banana

· DeepMind Translated
DeepMind

June 30, 2026

We are making it easier for you to experiment with and scale your creative ideas with Nano Banana 2 Lite—our fastest, most cost-effective Gemini image model—and Gemini Omni Flash for high-quality video generation and conversational editing.

Alisa Fortin

Product Manager, Google DeepMind

Anish Nangia

Product Manager, Google DeepMind

Today, we are making it faster and simpler to experiment, refine, and scale your ideas with two major releases:

Building generative media is often about creative iteration. With these two models, developers can build complete end-to-end multimedia experiences, connecting rapid image generation with video creation and editing. Whether your workflow requires generating thousands of images or editing multi-turn video sequences, you now have two new models to build faster, iterate seamlessly, and bring your creative vision to life.

Nano Banana 2 Lite: Our Fastest, Most Cost-Effective Gemini Image Model

Watch a side-by-side comparison of Nano Banana 2 Lite and Nano Banana 2 in speed and quality when generating images using simple prompts.

Nano Banana 2 Lite (gemini-3.1-flash-lite-image) is designed for rapid creative ideation and high-speed development pipelines where speed and cost are the primary constraints. For developers currently using our first-generation Nano Banana (gemini-2.5-flash-image), we recommend replacing it with Nano Banana 2 Lite; you can switch now and see immediate benefits across key performance dimensions.

Performance benchmarks comparing Nano Banana 2 and 2 Lite against competitor AI image models, evaluating the trade-offs between generation/editing quality (Elo score), processing latency, and cost per 1,000 resolution images.

Key Advantages of Nano Banana 2 Lite:

  • Latency: Delivers text-to-image outputs in under 4 seconds. This makes it ideal for interactive prototyping and rapid visual sketching.
  • Cost-effectiveness ($0.034 per 1,000 images): A highly cost-effective choice for developers focused on drafting, creative ideation, managing operational budgets, or low-bandwidth usage.

Despite prioritizing speed, Nano Banana 2 Lite still maintains reliable prompt adherence, strong character consistency, and clear, readable text rendering within images.

Understanding the Nano Banana Family

  • Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image): Built for speed. Optimized for near-real-time, high-volume workflows, excelling in scenarios where ultra-low latency is critical.
  • Nano Banana 2 (Gemini 3.1 Flash Image): The versatile workhorse. Delivers high-quality outputs with lower latency, achieving the optimal balance between performance and cost.
  • Nano Banana Pro (Gemini 3 Pro Image): Optimized for complex, professional-grade use cases. It offers the strongest control and advanced reasoning capabilities for tasks where accuracy is more important than speed.
  • Nano Banana (Gemini 2.5 Flash Image): Our legacy model. We recommend upgrading to Nano Banana 2 Lite for better quality, faster speeds, and lower costs.

For a complete list of model capabilities and integration guides, please refer to the developer documentation.

Alongside its release on developer platforms, Nano Banana 2 Lite is also coming to Google’s consumer products, including AI Mode in Search, the Gemini app, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads.

Experience High-Quality, Cost-Effective Video Editing and Generation with Gemini Omni Flash

Watch a person perform four digital magic tricks using Gemini Omni, such as pulling a 3D balloon word out of a phone and pouring water from a screen into a glass. In the corner, a small “original” video shows how she actually filmed these magic effects before adding the Omni-generated special effects.

At Google I/O, we introduced Gemini Omni Flash, a model that combines Gemini’s multimodal reasoning with video generation and editing. Today, Gemini Omni Flash (gemini-omni-flash-preview) is rolling out to developers via the Gemini API and Google AI Studio, natively supporting high-quality video generation and conversational editing from combinations of text, image, and video inputs. The model is competitively priced at $0.10 per second of video output, matching Veo 3.1 Fast.

Key Advantages of Omni Flash:

  • Conversational video editing: Refine and edit videos using natural language.
  • Multimodal referencing: Combine inputs like images, text, and video to maintain control and consistency over the scene.
  • Real-world knowledge: Omni leverages Gemini’s knowledge—such as history, biology, and narrative logic—to construct engaging videos.
  • Text and action synchronization: Link text and graphics directly to video actions with simple prompts.

For comprehensive benchmarking information, please visit Google DeepMind’s Gemini Omni page.

Limitations:

  • Currently, Omni only supports 10-second video generation, with longer durations coming soon.
  • The model does not yet support audio reference uploads and scene expansion in the Gemini API.
  • The API architecture accepts video references up to 3 seconds in duration.