Nano Banana 2 (Gemini 3.1 Flash Image Preview)

google/gemini-3.1-flash-image-preview
VisionJSONReasoning
by Google · 2026-02-26

Gemini 3.1 Flash Image Preview, a.k.a. "Nano Banana 2," is Google’s latest state of the art image generation and editing model, delivering Pro-level visual quality at Flash speed. It combines...

ctx65.5K tokens
Inputimage + text
Outputimage + text
p50 TTFT10.00 s
INPUT/ 1M tokens
OUTPUT/ 1M tokens
p50 TTFT10.00 s7d
p95 TTFT10.00 s7d
TRAFFIC172.6Ktokens / 7d

Google: Nano Banana 2 is a multimodal model that processes both images and text input. It is based on the Gemini 3.1 Flash architecture, which emphasizes fast inference. The model accepts image…

What is Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)?

Who is this model designed for?

How do image and text input work together?

Code samples

Call from any SDK

  • Gemini SDKhttps://api.orcarouter.ai

Supported parameters

  • include_reasoning
  • max_tokens
  • reasoning
  • response_format
  • seed
  • stop
  • structured_outputs
  • temperature
  • top_p

Pricing

Per request$0.1510
CurrencyUSD
Flat fee per API call (image generation models)

Performance

p50 TTFT
10.00 s
Output speed
2401 tok/s
p95 TTFT
10.00 s
Error rate
0.78%

Public benchmarks

51.8
AA Coding
Better than 72% of models compared
#30 of 106
53.8
AA Intelligence
Better than 72% of models compared
#31 of 110
53.8
AA Math
Better than 30% of models compared
#57 of 81
GPQA Diamond
50.8 index
MMLU-Pro
58.8 index
τ²-Bench
40.8 index
Source: artificialanalysis.ai

FAQ

How is pricing determined for this model on OrcaRouter?
Pricing is per token (input + output), with image tokens counted. For exact rates, see OrcaRouter’s pricing page. Costs vary by usage volume.
What is the maximum context window size?
The context window is 65,536 tokens, including both text and encoded image tokens.
What are the main strengths of this model?
It combines image understanding with text generation in a single, fast flash architecture, suitable for real-time multimodal applications.
How does it compare to GPT-4V or Claude 3 Vision?
It is generally faster and likely cheaper, but may have lower accuracy on complex visual reasoning tasks. No direct benchmarks are available.
Does OrcaRouter handle data privacy when using this model?
OrcaRouter's data handling policies should be reviewed. Typically, API calls are processed through Google's backend; data may be retained according to Google's terms. Consult OrcaRouter’s privacy documentation.
How do I call this model using an OpenAI-compatible API?
Use base_url https://api.orcarouter.ai/v1, with model id "google/gemini-3.1-flash-image-preview". Send a POST request to /chat/completions with messages including image_url objects.
What image formats are supported?
Common formats like JPEG and PNG are likely supported. Base64 encoding or public URLs are accepted. Exact format restrictions are not documented.
Can I use this model without images?
Yes, you can send text-only prompts. However, the model is designed for multimodal input; pure text performance may be similar to the standard Flash model.
Are there any known limitations regarding safety filters?
Google applies safety filters to inputs and outputs. This may block certain images or text. OrcaRouter may also enforce content policies. Test with your content.
How can I estimate token usage for images?
Image tokens depend on resolution and detail. No fixed formula is provided. Monitor the prompt_tokens in the API response to gauge usage per request.

Embed this badge

Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)pricing pending10000ms p50via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/google/gemini-3.1-flash-image-preview" target="_blank"> <img src="https://www.orcarouter.ai/embed/google/gemini-3.1-flash-image-preview.svg" alt="Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview) on OrcaRouter" /> </a>
Markdown [![Google: Nano Banana 2 (Gemini 3.1 Flash Image Preview)](https://www.orcarouter.ai/embed/google/gemini-3.1-flash-image-preview.svg)](https://www.orcarouter.ai/models/google/gemini-3.1-flash-image-preview)

Model card as data

GET /api/public/models/google/gemini-3.1-flash-image-previewOpen
Machine-readable:/llms.txt/llms-full.txt