Choose Auto, TokenLab Verified, or Official for each request, with prices shown up front. See what's new

Google: Nano Banana 2.1

Nano Banana 2.1 is the newer version of Nano Banana 2 for creating and revising images from a prompt and reference pictures. It holds a subject steady over several rounds of changes and renders text more cleanly, which helps when a set of images must share one look.
Compare models
nano-banana-2.1
AvailableGoogleImageSync / Async
Price
From $0.0168
Released
Oct 6, 2026
Modalities
Image

About Nano Banana 2.1

Nano Banana 2.1 is Google's successor to Nano Banana 2, released in October 2026 for creating and revising images from prompts and reference pictures. Google now treats it as the default Flash image model for new work. Compared with Nano Banana 2, subjects hold steadier across rounds of edits, prompts are followed more closely, and wide banners at 2K and 4K no longer break into repeated tiles.

Where it works well

  • Keeps a character recognizable across several rounds of edits better than Nano Banana 2, per Google.
  • Follows written prompts more closely than Nano Banana 2, according to Google's documentation.
  • Banner shapes such as 8:1 and 1:4 render at 2K and 4K without the tiling seams of the earlier version.
  • Cleaner lettering makes short titles and labels more usable in finished campaign images.
  • The model card names creating and editing images over many quick iterations as an intended use.

When to choose another model

  • Nano Banana Pro remains Google's choice for the hardest compositions that need deep world knowledge.
  • Fine print and paragraph-length text are still weak points listed in Google's model card.
  • Edits can keep the pose of the input subject, and 3D reasoning is limited, per the model card.

Getting started

  1. Create API key

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

  2. 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
    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": "nano-banana-2.1",
      "prompt": "A minimalist product photo of matte black headphones on a soft blue background."
    }'

Parameters

Image editing

ParameterValues and limitsDefaultNotes
aspect_ratio1:1, 1:4, 1:8, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 8:1, 9:16, 16:9, 21:9Not providedNot provided
asyncNot providedNot providedNot provided
imageNot providedNot providedThis operation accepts image inputs. The maximum total source-image count is not yet verified for this public operation.
image_urlNot providedNot providedNot provided
image_urlsNot providedNot providedNot provided
operationNot providedNot providedNot provided
promptNot providedNot providedNot provided
reference_image_urlsNot providedNot providedNot provided
resolution1k, 2k, 4kNot providedNot provided
response_formatNot providedNot providedThe public image response selector accepts url or b64_json. A URL-only supply rejects b64_json before upstream execution. Omit this field to keep the selected supply response representation; no uniform omitted-field default is declared here.
sizeNot providedNot providedUse the declared resolution and aspect_ratio to select the corresponding output size. size follows the model-specific resolution table; arbitrary pixel dimensions are not declared as a public control.

Image to image

ParameterValues and limitsDefaultNotes
aspect_ratio1:1, 1:4, 1:8, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 8:1, 9:16, 16:9, 21:9Not providedNot provided
asyncNot providedNot providedNot provided
imageNot providedNot providedThis operation accepts image inputs. The maximum total source-image count is not yet verified for this public operation.
image_urlNot providedNot providedNot provided
image_urlsNot providedNot providedNot provided
operationNot providedNot providedNot provided
promptNot providedNot providedNot provided
reference_image_urlsNot providedNot providedNot provided
resolution1k, 2k, 4kNot providedNot provided
response_formatNot providedNot providedThe public image response selector accepts url or b64_json. A URL-only supply rejects b64_json before upstream execution. Omit this field to keep the selected supply response representation; no uniform omitted-field default is declared here.
sizeNot providedNot providedUse the declared resolution and aspect_ratio to select the corresponding output size. size follows the model-specific resolution table; arbitrary pixel dimensions are not declared as a public control.

Text to image

ParameterValues and limitsDefaultNotes
aspect_ratio1:1, 1:4, 1:8, 2:3, 3:2, 3:4, 4:1, 4:3, 4:5, 5:4, 8:1, 9:16, 16:9, 21:9Not providedNot provided
asyncNot providedNot providedNot provided
operationNot providedNot providedNot provided
promptNot providedNot providedNot provided
resolution1k, 2k, 4kNot providedNot provided
response_formatNot providedNot providedThe public image response selector accepts url or b64_json. A URL-only supply rejects b64_json before upstream execution. Omit this field to keep the selected supply response representation; no uniform omitted-field default is declared here.
sizeNot providedNot providedUse the declared resolution and aspect_ratio to select the corresponding output size. size follows the model-specific resolution table; arbitrary pixel dimensions are not declared as a public control.

Use this model for text-to-image generation.

Pricing

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

Text

per image
Official price
$1.50/image
Verified price
$0.75/image
Discount
-50%

Text to image / Image to image / Image editing · 1k

per image
Official price
$0.0336
Verified price
$0.0168
Discount
-50%

Text to image / Image to image / Image editing · 2k

per image
Official price
$0.0504
Verified price
$0.0252
Discount
-50%

Text to image / Image to image / Image editing · 4k

per image
Official price
$0.113
Verified price
$0.0565
Discount
-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
30-day success rate
100.0%
30-day requests
100+
Last active
2 weeks ago

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

Open in Console

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

Help me create an image with 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
  • Character series

    Draw one mascot or character in several scenes over a few turns, sending the earlier image as reference so the look carries over.

  • Upgrading Nano Banana 2 flows

    Change the model ID in an existing Nano Banana 2 pipeline, then recheck a few saved prompts before you switch all traffic.

  • Campaign image sets

    Produce matching square, story, and banner versions of one visual so that a launch set shares the same subject and style.

  • Iterative design rounds

    Refine a draft through short follow-up instructions, such as moving an object or changing the light, instead of rewriting the full prompt.

Prompt examples

Using the attached mascot, draw him surfing a big wave at sunset, same colors and outfit, 16:9.

Now move the surfboard to the left side and make the sky stormy, keep everything else the same.

A 1:4 vertical poster of a mountain trail at dawn with the title "Summit Week" at the top, 2K.

FAQ

What is Nano Banana 2.1?

It is Google's update to Nano Banana 2, released in October 2026. It creates images from prompts, turns reference pictures into new scenes, and revises supplied images from instructions, with 1K, 2K, or 4K output and fourteen aspect ratios.

Should I move from Nano Banana 2 to 2.1?

For new projects, yes: Google recommends Nano Banana 2.1 and calls Nano Banana 2 the previous generation. Existing prompts generally carry over because the request style and aspect ratios match, but results will look somewhat different, so recheck prompts tuned to a specific look.

Is Nano Banana 2.1 the same as gemini-nano-banana-2.1?

Yes, the two IDs lead to one Google model under two names. The catalog keeps them as separate entries with their own request details and price tables, so read both pages and pick the one that fits your integration.

Can it handle long text in images?

Short headings and labels render more cleanly than in Nano Banana 2. Google's model card still lists blurry small text and poor rendering of long paragraphs as limitations, so keep on-image copy brief and set longer text in your layout tool.

How much does Nano Banana 2.1 cost?

On TokenLab, Nano Banana 2.1 costs $0.0168-$0.0565 per image. 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 Nano Banana 2.1 use?

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

Which operations does Nano Banana 2.1 support?

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

Compare Nano Banana 2.1

Sources

Reviewed Oct 8, 2026

More from Nano Banana

Related models