Google: Gemini Nano Banana 2.1

Gemini Nano Banana 2.1 is Google's update to its Flash image model for generating images and editing them in conversation. Text inside images is clearer and characters and objects stay steadier across edits, which suits infographics, menus and product shots that need readable lettering in 1K, 2K or 4K output.

Compare models

gemini-nano-banana-2.1

AvailableGoogleImageSync / AsyncNew

Input / Output-50%: $1.50 / $7.50$0.75 / $3.75
Context: 128K
Released: Oct 6, 2026
Max output: 32K
Modalities: ImageVision

About Gemini Nano Banana 2.1

Gemini Nano Banana 2.1 is Google's October 2026 update to Gemini 3.1 Flash Image, a Flash-tier model for generating pictures and editing them through conversation. Google calls it its main high-efficiency image model. It points to clearer text rendering, more accurate infographic layouts, and better realism at 1K, 2K, and 4K, while the model keeps Flash-level speed.

Where it works well

Text inside images and infographic layouts come out more accurately than in Gemini 3.1 Flash Image, according to Google.
Google fixed the tiling artifacts that wide and panoramic formats such as 4:1 and 8:1 showed at 2K and 4K.
Characters stay closer to their earlier look when you revise one image over several turns.
Google's model card lists rendering localized text in several languages among its intended uses.

When to choose another model

Google's model card notes that small text can blur, especially at 1K, and that long paragraphs render poorly.
For the most complex scenes and precise creative control, Google points to Nano Banana Pro instead.
Characters do not always match the input exactly, and left and right positions are sometimes confused.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Image editingImage to imageText to imageTokenLab endpoint

POST/v1/images/generations

API format:

curl -X POST "https://api.tokenlab.sh/v1/images/generations" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "aspect_ratio": "1:1",
  "operation": "text-to-image",
  "resolution": "1k",
  "model": "gemini-nano-banana-2.1",
  "prompt": "A minimalist product photo of matte black headphones on a soft blue background."
}'

Pricing

TokenLab price applies to Verified, which costs less on most models. Official price is the model maker's published price and applies to the more reliable Official route. Auto charges the route that completes the request.

Text

per 1M tokens

Official price: Input $1.50 / Output $7.50 / Text Input $1.50 / Image Input $1.50 / Text Output $7.50
TokenLab price: Input $0.75 / Output $3.75 / Text Input $0.75 / Image Input $0.75 / Text Output $3.75
Discount: -50%

Image

per 1M tokens

Official price: Input $1.50 / Output $30.00 / Text Input $1.50 / Image Input $1.50 / Text Output $7.50 / Image Output $30.00
TokenLab price: Input $0.75 / Output $15.00 / Text Input $0.75 / Image Input $0.75 / Text Output $3.75 / Image Output $15.00
Discount: -50%

	Official priceper 1M tokens	TokenLab priceper 1M tokens	Discount
Text	Input $1.50 / Output $7.50 / Text Input $1.50 / Image Input $1.50 / Text Output $7.50	Input $0.75 / Output $3.75 / Text Input $0.75 / Image Input $0.75 / Text Output $3.75	-50%
Image	Input $1.50 / Output $30.00 / Text Input $1.50 / Image Input $1.50 / Text Output $7.50 / Image Output $30.00	Input $0.75 / Output $15.00 / Text Input $0.75 / Image Input $0.75 / Text Output $3.75 / Image Output $15.00	-50%

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 8, 2026, 08:57 AM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Gemini Nano Banana 2.1 in Console with a prompt ready to edit or send.

Help me create an image with gemini-nano-banana-2.1 using text-to-image at /v1/images/generations. Show the result, cost, and any limits I should know.

Use cases

Best for

Text to Image
Vision

Infographics and diagrams
Lay out a labeled process diagram or a data-style infographic where headings and short labels must stay legible at 2K or 4K.
Menus and signage
Produce a café menu board or a shop sign with short item names, then correct wording or colors in a follow-up edit.
Product shot revisions
Send a product photo with an instruction to change the backdrop, props, or lighting while the product itself stays as it is.
Panoramic headers
Generate 4:1 or 8:1 website headers and event banners at 2K or 4K, where the earlier model could show repeated tiles.

Prompt examples

A clean infographic titled "How coffee is roasted" with five numbered steps, short labels and flat icons, 3:4, 2K.

Edit this product photo: replace the grey background with a sunlit marble counter and keep the bottle and its label unchanged.

An 8:1 panoramic header of a night market street lit by paper lanterns, 4K, no text.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What is Gemini Nano Banana 2.1?

It is Google's Flash-tier image model released in October 2026 as the update to Gemini 3.1 Flash Image (Nano Banana 2). It generates images from text, builds new images from reference pictures, and edits supplied images from written instructions, with 1K, 2K, or 4K output.

How is it different from Gemini 3.1 Flash Image?

Google lists better visual quality and realism, clearer text rendering, more accurate infographic layouts, closer prompt adherence, and steadier characters over several edits. It also removes tiling artifacts in wide and panoramic images at 2K and 4K. Google now recommends it for new projects.

Is gemini-nano-banana-2.1 the same model as nano-banana-2.1?

Yes. Both catalog IDs reach Google's Nano Banana 2.1; this one carries the Gemini name and the other the Nano Banana name. They are separate catalog entries, so compare the request details and the price table on each page before you choose.

Do generated images carry a watermark?

Yes. Google states that all images generated by its Gemini image models include a SynthID watermark. Plan for this in any workflow where generated images must be labeled, tracked, or checked later.

When should I use Nano Banana Pro instead?

Pick Pro for the most complex visual tasks, where Google cites its deeper world knowledge and more precise creative control. Stay with this model for everyday generation and editing at Flash speed, including text-heavy graphics that do not need that extra control.

How much does Gemini Nano Banana 2.1 cost?

On TokenLab, Gemini Nano Banana 2.1 costs Input $0.75 / Output $3.75 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Gemini Nano Banana 2.1 use?

Use https://api.tokenlab.sh/v1/images/generations for Gemini Nano Banana 2.1. The request example below shows the matching code shape.

Which operations does Gemini Nano Banana 2.1 support?

Gemini Nano Banana 2.1 supports Image editing, Image to image, Text to image. Select an operation above to see its endpoint and request example.