gemini-3.1-flash-lite-image ~4s generation

Nano Banana 2 Lite
from idea to image, without the wait

Nano Banana 2 Lite is Google's fastest, most cost-efficient image model β€” built for rapid ideation, character consistency, and natural language editing right inside Google AI Studio.

One-shot editing Multi-image support Natural language
Try:
Start from an example

Your image will appear here

Powered by Nano Banana 2 Lite
~4s Text-to-image latency
$0.034 Per 1K resolution image
2.7Γ— Faster than Nano Banana 2
SynthID Invisible AI watermark

The fastest image model in the Nano Banana family

Built on Gemini 3.1 Flash-Lite Image, Google Nano Banana 2 Lite delivers high-speed generation and editing at Google's lowest cost yet β€” without compromising the character consistency and scene preservation you expect from Nano Banana. It ships in Google AI Studio alongside the wider Gemini Omni Flash release.

It's the recommended replacement for the legacy Nano Banana (gemini-2.5-flash-image), offering better quality, faster speeds, and lower costs for developers, creators, and businesses.

Try in Google AI Studio β†’
Nano Banana 2 Lite
  • ~4s latency
  • $0.034 / image
  • High throughput
Recommended
Nano Banana 2
  • Balanced quality
  • Generalist workhorse
  • Best quality/cost ratio
Nano Banana Pro
  • Complex tasks
  • Advanced reasoning
  • Professional control

Built for rapid-fire visual exploration

Lightning-fast latency

Text-to-image outputs in ~4 seconds. Ideal for interactive prototyping and real-time apps where speed is critical.

Cost-efficient at scale

Generate thousands of images at $0.034 each β€” a fraction of the cost of heavier production models.

Character consistency

Maintain consistent characters across multiple edits β€” perfect for storytelling, game design, and branded content.

Scene preservation

Keeps original structure, perspective, and lighting intact while applying precise, natural-looking transformations.

Legible text rendering

Despite prioritizing speed, retains reliable prompt adherence and clear in-image text rendering.

Multi-image workflow

Process and edit multiple images with consistent styling β€” ideal for batch editing and series creation.

From idea to image in seconds

See how developers and creators are building with Nano Banana 2 Lite across real-world applications.

Speed without sacrificing quality

Nano Banana 2 Lite offers the best balance of quality and speed in the Nano Banana family.

Nano Banana 2 Lite result
Nano Banana 2 result
⟺
Nano Banana 2 Nano Banana 2 Lite

Prompt: A lone astronaut stands poised at the threshold of a futuristic corridor, gazing out towards a distant, mountainous landscape.

Latency (1K image) ~4s
Cost per image $0.034
Image quality (Elo) High

Speed & cost, visualized

Same prompt, three models. Watch Nano Banana 2 Lite finish first β€” and stay cheapest β€” while heavier models are still rendering.

Generating…
0.0s
Nano Banana 2 Lite gemini-3.1-flash-lite-image β€”
βœ“
Nano Banana 2 gemini-3.1-flash-image β€”
βœ“
Typical competitor heavy production model β€”
βœ“
Cost to generate 1,000 images
2 Lite $0
Nano Banana 2 $0
Competitor $0

β‰ˆ $0.034 per 1K-resolution image β€” a fraction of heavier production models.

Edit images smarter with Nano Banana 2 Lite

01

Upload or describe

Start with a text prompt for generation, or upload any image you want to edit. Works with photos, artwork, or any visual content.

02

Describe your edit

Write a clear text prompt describing your transformation. Intelligent prompt understanding ensures edits align with your creative vision.

03

Download your result

Get your professionally edited, high-resolution image in seconds. Production-ready results with SynthID watermarking for transparency.

Trusted by creators and developers

"Speed is no longer a limitation. When generation is faster than imagination, creators can stay inside the idea instead of waiting on the tool."
Idan Yonas β€” Director of AI Content, Artlist
"Nano Banana 2 Lite is fast and reliable, helping designers explore more ideas to craft unique images. Ideal for rapid iteration while staying in the creative flow."
Itay Schiff β€” Co-founder, Figma Weave
"Its speed suits real-time scenarios well, allowing our AI Agent to iterate on visuals quickly and deliver results in seconds. Image quality comes close to the full Nano Banana 2."
Tao Zhang β€” Co-Founder & CPO, Manus AI
"Instant-ramen delivers consistent, high-quality 1k images ~2.7Γ— faster with incredibly tight latency variance. It's what makes real-time generative play viable at scale."
Max Child β€” CEO, Weekend

Choose the plan that fits your needs

Start free in demo mode. Upgrade for more credits, higher resolution, and Pro-model access.

Starter

Billed monthly
$20 /mo
400 Credits/mo Up to 80 images / month
  • Download image files
  • HD image resolution
  • Use of Basic AI model
  • Standard generation quality
  • 1 running job at once
  • No ads
  • No watermark
Best value

Premium

Best value
$160 $80 /mo
2,000 Credits/mo Up to 400 images / month
  • Download image files
  • HD image resolution
  • Use of Basic & Pro AI models
  • Highest generation quality
  • Private generations
  • 8 running jobs at once
  • Priority generation queue
  • Advanced image editing
  • Batch processing
  • Early access to new features
  • No ads
  • No watermark

Frequently asked questions

What is Nano Banana 2 Lite?

Nano Banana 2 Lite (gemini-3.1-flash-lite-image) is Google's fastest, most cost-efficient image model in the Nano Banana family. It delivers text-to-image generation in ~4 seconds at $0.034 per 1K resolution image, while maintaining character consistency and scene preservation.

How does it differ from Nano Banana 2?

Nano Banana 2 Lite prioritizes speed and cost-efficiency for high-volume workflows, while Nano Banana 2 (gemini-3.1-flash-image) is the generalist workhorse offering the best balance of quality and cost. For complex professional tasks, Nano Banana Pro provides the most robust control.

Can I use it for professional projects?

Yes. Nano Banana 2 Lite is used by companies like Artlist, Figma Weave, Manus AI, and Latitude for real-time image generation in production workflows β€” from slide decks and web pages to living visual game worlds.

What edits does it support?

Text-to-image generation, image editing with natural language prompts, multi-image composition, character consistency across edits, style transformations, and in-image text rendering.

How fast is image processing?

Text-to-image outputs typically arrive in ~4 seconds. In benchmarks, it's ~2.7Γ— faster than Nano Banana 2 (Gemini 3.1 Flash Image) with tight latency variance β€” ideal for real-time applications.

Can I use Nano Banana 2 Lite in an AI agent workflow?

Absolutely. Many teams orchestrate Nano Banana 2 Lite inside agents alongside a reasoning model such as Claude Sonnet 5 β€” Sonnet 5 plans and refines the prompt, then Nano Banana 2 Lite renders the visuals in seconds. You can also hand stills to Google Flow (the Flow AI filmmaking tool) to turn them into video.

Where do I access it β€” Google AI Studio or the API?

Nano Banana 2 Lite is available in Google AI Studio for hands-on prompting and in the Gemini API for production. It launched as part of Google's Gemini Omni Flash family, so if you already use AI Studio you can switch to Nano Banana 2 Lite without leaving your workflow.

Are images watermarked?

Yes. All images include SynthID, Google's invisible digital watermark that allows AI-generated content to be identified. This supports safety and transparency across the web.

Start building with Nano Banana 2 Lite

The fastest path from prompt to production β€” available in Google AI Studio, Gemini API, and Gemini Enterprise.