Nano Banana Pro (Gemini 3 Pro Image Preview) pricing and routing profile
Nano Banana Pro (Gemini 3 Pro Image Preview) from Google lists at $2.00 per million input tokens and $12.00 per million output tokens, with a 65,536-token context window. Prices from OpenRouter's catalogue on 2026-09-28.
Structured outputReasoning control
Cost per 1,000 requests
| Request shape | List cost |
|---|---|
| Classification or routing (600 tokens in, 10 out) | $1.32 |
| Chat reply (2,000 in, 400 out) | $8.80 |
| Retrieval-augmented answer (8,000 in, 500 out) | $22.00 |
Before prompt caching. Try your own shape in the cost calculator.
Can Gimbix route to it first?
Not as a first-answer model through OpenRouter: no host returns token log probabilities for Nano Banana Pro (Gemini 3 Pro Image Preview). It can still be the model Gimbix falls back to when a cheaper model is unsure.
Cheaper models that can answer first
Models priced below Nano Banana Pro (Gemini 3 Pro Image Preview) that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.
| Model | Input / output per 1M | Cheaper on a chat reply |
|---|---|---|
| Grok 4.7 | $1.60 / $4.80 | 42% |
| Qwen3.7 Max | $1.48 / $4.42 | 46% |
| GLM 5.1 | $1.40 / $4.40 | 48% |
| Qwen3.6 Max Preview | $1.03 / $6.16 | 49% |
| Grok 4.3 | $1.25 / $2.50 | 60% |
| Grok 4.20 Multi-Agent | $1.25 / $2.50 | 60% |
Questions
How much does Nano Banana Pro (Gemini 3 Pro Image Preview) cost?
Nano Banana Pro (Gemini 3 Pro Image Preview) lists at $2.00 per million input tokens and $12.00 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $8.80 per 1,000 requests.
What is the context window of Nano Banana Pro (Gemini 3 Pro Image Preview)?
65,536 tokens.
Can a cheaper model replace Nano Banana Pro (Gemini 3 Pro Image Preview)?
Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match Nano Banana Pro (Gemini 3 Pro Image Preview) on one half of your requests and make at most three answers worse on the other.