Nano Banana 2 Lite and Gemini Omni Flash Release

Nano Banana 2 Lite and Gemini Omni Flash Release

Google DeepMind has released Nano Banana 2 Lite and Gemini Omni Flash, providing developers with high-speed image generation and cost-efficient video creation and editing tools. These models enable the creation of end-to-end multimedia experiences by combining rapid visual drafting with cinematic video animation.

Nano Banana 2 Lite: High-Throughput Image Generation

Nano Banana 2 Lite (gemini-3.1-flash-lite-image) is optimized for rapid ideation and high-velocity developer pipelines where speed and cost are the primary constraints. It is the recommended replacement for the legacy Nano Banana (gemini-2.5-flash-image) model.

Key Performance Metrics

  • Latency: The model delivers text-to-image outputs in 4 seconds.
  • Cost-Efficiency: Pricing is set at $0.034 per 1K-resolution image.

Despite the focus on speed, the model maintains reliable prompt adherence, legible in-image text rendering, and strong character consistency.

The Nano Banana Model Family

  • Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image): Optimized for ultra-low latency and high-volume workflows.
  • Nano Banana 2 (Gemini 3.1 Flash Image): A generalist model balancing performance and cost.
  • Nano Banana Pro (Gemini 3 Pro Image): Designed for professional use cases requiring advanced reasoning and high accuracy.
  • Nano Banana (Gemini 2.5 Flash Image): Legacy model; users are encouraged to upgrade to Nano Banana 2 Lite.

Nano Banana 2 Lite is available via Google AI Studio, the Gemini API, and the Gemini Enterprise Agent Platform, and is rolling out to consumer surfaces including the Gemini app, AI Mode in Search, NotebookLM, Google Photos, Stitch, Google Flow, and Google Ads.

Gemini Omni Flash: Conversational Video Generation and Editing

Gemini Omni Flash (gemini-omni-flash-preview) integrates multimodal reasoning with video generation. It is now available to developers via the Gemini API and Google AI Studio, as well as the Gemini app and Google Flow.

Core Capabilities

  • Conversational Editing: Users can refine and edit videos using natural language.
  • Multimodal Referencing: The model combines text, image, and video inputs to maintain scene consistency.
  • Knowledge Integration: Omni Flash utilizes Gemini's internal knowledge of history, biology, and narrative logic to construct videos.
  • Action Synchronization: Text and graphics can be connected directly to video actions through prompting.

Pricing and Technical Limitations

Gemini Omni Flash is priced at $0.10 per second of video output, matching the cost of Veo 3.1 Fast. Current limitations include:

  • Duration: Video generations are currently limited to 10 seconds.
  • Unsupported Features: Audio references and scene extension are not yet supported in the Gemini API.
  • Processing Issues: Video references up to 3 seconds are accepted by the API schema but are not currently processed correctly by the model.
  • Consistency: There are existing limitations regarding character consistency during panning movements or scene changes.

Integrated Multimedia Workflows

Developers can chain Nano Banana 2 Lite and Gemini Omni Flash to create seamless pipelines. A common workflow involves using Nano Banana 2 Lite for high-speed image generation and then passing that image as a reference to Gemini Omni Flash to animate it into a video.

Using the Interactions API, developers can maintain session history and context, allowing users to perform up to three sequential edits.

Demo Applications

  • Anywhere: Uses Nano Banana 2 Lite to transport users to landmarks via photos, then uses Omni Flash to animate those images into clips.
  • Space Lift: An interior design app that generates room concepts with Nano Banana 2 Lite and creates cinematic showcases with Omni Flash.
  • Omni Product Studio: Converts static e-commerce images created by Nano Banana 2 Lite into cinematic videos via Gemini Omni.

Safety and Transparency

Both Gemini Omni and Nano Banana 2 Lite utilize SynthID watermarking to ensure AI-generated content can be verified through the Gemini app, Gemini in Chrome, or Search.

Sources