Google: Gemini Nano Banana 2 Lite
Nano Banana 2 Lite is Google's low-cost image generation and editing model in the Gemini API, with the id gemini-3.1-flash-lite-image. According to the Gemini API release notes, Google released it to general availability on June 30, 2026 as a multimodal model for low-latency, cost-effective image generation and editing. The image generation guide calls it the fastest and cheapest Gemini image model, aimed at high-volume work where speed and cost matter most.
Model Specs
| Spec | Nano Banana 2 Lite |
|---|---|
| Model id | gemini-3.1-flash-lite-image |
| General availability | June 30, 2026 |
| Output resolution | 1K only |
| Aspect ratios | 1:1, 3:2, 2:3, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9 |
| Thinking levels | minimal (default) and high |
| Reference images | Up to 14 object images; no character or style references |
| Search grounding | Not supported |
Google's guide also says the model supports video-to-image generation and is not optimized for multiple reference inputs or multi-turn sequential editing.
Pricing
Google's pricing page, checked 2026-10-29, lists Standard rates per 1M tokens for Gemini 3.1 Flash Lite Image (Nano Banana 2 Lite):
| Component | Price |
|---|---|
| Input (text, image, video) | $0.25 |
| Output (text and thinking) | $1.50 |
| Output (images) | $30.00 |
The page's footnote says a 1K (1024x1024) output image uses 1,120 tokens, which works out to $0.0336 per image.
How It Compares
According to the same guide, Nano Banana 2 (gemini-3.1-flash-image) adds 512px, 2K and 4K output, Google Search grounding, and support for character images, while Lite stays at 1K. Nano Banana Pro (gemini-3-pro-image) is described as the premium model for the most complex tasks, with more reference types. The guide recommends Nano Banana 2.1 (gemini-nano-banana-2.1) for new projects, so Lite is the choice when per-image cost and speed matter more than resolution or editing depth.
Use Cases
- Bulk generation of thumbnails, product mockups and social images at 1K.
- Fast prototyping of image prompts before moving to a higher tier.
- Pipelines that feed a few object reference images and need low latency.
A limitation: 1K only, no search grounding, and weaker support for multi-image or multi-turn editing, so complex edits belong on a higher Nano Banana tier.
Getting Started
- Get a Gemini API key from Google AI Studio.
- Call
gemini-3.1-flash-lite-imagewith a text prompt, and optionally up to 14 object images. - Start with the default
minimalthinking level, and tryhighonly if prompts need more reasoning. - Multiply expected volume by roughly $0.0336 to estimate image cost.
FAQ
What is Nano Banana 2 Lite?
Google's cheapest and fastest Gemini image generation and editing model, id gemini-3.1-flash-lite-image.
What resolution does it support?
Only 1K, according to Google's image generation guide.
How much does it cost per image?
Google's pricing footnote gives $0.0336 for a 1K image, based on $30 per 1M image output tokens.
Is it still available?
Yes. It reached general availability on June 30, 2026, and the deprecations page lists no shutdown date for it.
Alternatives
- GPT Image 2: OpenAI's image model.
- FLUX.1 Pro: an earlier Black Forest Labs image model.
- Qwen Image 2.1: Alibaba's image model.
- Gemini 3.1 Pro: Gemini's text-output Pro model, for prompt writing and analysis.
Conclusion
Nano Banana 2 Lite suits teams that generate many 1K images and want the lowest Gemini price and latency. Check the image generation guide for the current lineup, and see more in the models category.
Sources
- Websitehttps://ai.google.dev/gemini-api/docs/image-generation
- Social Mediahttps://x.com/GoogleAI
Comments
No comments yet. Be the first to comment!