Gemini API pricing, every Google model per million tokens
On 10 October 2026, Google models on this list cost from $0.05 per million input tokens (Gemma 3 4B) to $2.00 (Nano Banana Pro (Gemini 3 Pro Image)), and from $0.10 per million output tokens (Gemma 3 4B) to $12.00 (Nano Banana Pro (Gemini 3 Pro Image)).
| Model | Input | Output | Cached input | Batch in / out | Context | Your month | Listed |
|---|---|---|---|---|---|---|---|
| Nano Banana 2.1 | $1.50 | $7.50 | 66K | $202.50 | 2026-10-06 | ||
| Gemini 3.8 Flash promotional price, regular $1.50 in and $7.50 out | $0.75 | $3.75 | $0.075 | $0.375 / $1.88 | 1.05M | $101.25 | 2026-09-02 |
| Gemini 3.7 Flash promotional price, regular $1.50 in and $7.50 out | $0.75 | $3.75 | $0.075 | $0.375 / $1.88 | 1.05M | $101.25 | 2026-08-13 |
| Gemini 3.6 Flash | $0.75 | $3.75 | $0.075 | $0.375 / $1.88 | 1.05M | $101.25 | 2026-07-21 |
| Gemini 3.5 Flash Lite | $0.30 | $2.50 | $0.03 | $0.15 / $1.25 | 1.05M | $55.50 | 2026-07-21 |
| Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | $0.25 | $1.50 | 66K | $37.50 | 2026-06-30 | ||
| Nano Banana 2 (Gemini 3.1 Flash Image) | $0.50 | $3.00 | 131K | $75.00 | 2026-06-18 | ||
| Nano Banana Pro (Gemini 3 Pro Image) | $2.00 | $12.00 | $0.20 | 131K | $300.00 | 2026-06-18 | |
| Gemini 3.5 Flash | $1.50 | $9.00 | $0.15 | $0.75 / $4.50 | 1.05M | $225.00 | 2026-05-19 |
| Gemini 3.1 Flash Lite | $0.25 | $1.50 | $0.025 | $0.125 / $0.75 | 1.05M | $37.50 | 2026-05-07 |
| Gemma 4 26B A4B free tier on OpenRouter | $0.15 | $0.60 | 262K | $18.00 | 2026-04-03 | ||
| Gemma 4 31B DeepInfra, cheapest of 12 hosts, free tier on OpenRouter | $0.09 | $0.34 | $0.05 | 262K | $10.50 | 2026-04-02 | |
| Gemini 3.1 Flash Lite Preview | $0.25 | $1.50 | $0.025 | 1.05M | $37.50 | 2026-03-03 | |
| Nano Banana 2 (Gemini 3.1 Flash Image Preview) | $0.50 | $3.00 | 66K | $75.00 | 2026-02-26 | ||
| Gemini 3.1 Pro Preview Custom Tools | $2.00 | $12.00 | $0.20 | 1.05M | $300.00 | 2026-02-25 | |
| Gemini 3.1 Pro Preview | $2.00 | $12.00 | $0.20 | $1.00 / $6.00 | 1.05M | $300.00 | 2026-02-19 |
| Gemini 3 Flash Preview | $0.50 | $3.00 | $0.05 | $0.25 / $1.50 | 1.05M | $75.00 | 2025-12-17 |
| Nano Banana Pro (Gemini 3 Pro Image Preview) | $2.00 | $12.00 | $0.20 | 66K | $300.00 | 2025-11-20 | |
| Nano Banana (Gemini 2.5 Flash Image) | $0.30 | $2.50 | $0.03 | 33K | $55.50 | 2025-10-07 | |
| Gemini 2.5 Flash Lite retiring 2026-10-20 | $0.10 | $0.40 | $0.01 | $0.05 / $0.20 | 1.05M | $12.00 | 2025-07-22 |
| Gemini 2.5 Flash retiring 2026-10-20 | $0.30 | $2.50 | $0.03 | $0.15 / $1.25 | 1.05M | $55.50 | 2025-06-17 |
| Gemini 2.5 Pro retiring 2026-10-20 | $1.25 | $10.00 | $0.125 | $0.625 / $5.00 | 1.05M | $225.00 | 2025-06-17 |
| Gemini 2.5 Pro Preview 06-05 | $1.25 | $10.00 | $0.125 | 1.05M | $225.00 | 2025-06-05 | |
| Gemma 3 4B served by DeepInfra | $0.05 | $0.10 | 131K | $4.50 | 2025-03-13 | ||
| Gemma 3 12B NextBit, cheapest of 2 hosts | $0.05 | $0.15 | 131K | $5.25 | 2025-03-13 | ||
| Gemma 3 27B DeepInfra, cheapest of 5 hosts | $0.08 | $0.16 | 131K | $7.20 | 2025-03-12 | ||
| Gemma 2 27B served by NextBit | $0.65 | $0.65 | 8K | $48.75 | 2024-07-13 |
5 of these 27 models are not served by Google itself on OpenRouter, so their price is the cheapest regular price among the hosts that run them, and the host is named under the model. Compare Google with every other provider
Common questions
How much does the Gemini API cost?
On 10 October 2026, Google models on this list cost from $0.05 per million input tokens (Gemma 3 4B) to $2.00 (Nano Banana Pro (Gemini 3 Pro Image)), and from $0.10 per million output tokens (Gemma 3 4B) to $12.00 (Nano Banana Pro (Gemini 3 Pro Image)).
What is the cheapest Gemini model?
For a typical mix of three input tokens to every output token, the cheapest is Gemma 3 4B at $0.05 per million input tokens and $0.10 per million output tokens.
What is the newest Gemini model on the list?
Nano Banana 2.1, listed on 6 October 2026, at $1.50 per million input tokens and $7.50 per million output tokens.
Which Gemini model has the largest context window?
Gemini 3.8 Flash, with 1.05M tokens of context.
Does Google offer batch pricing?
Yes, 11 of its models on this list show a batch price, with batch input at a median of 50 percent of the standard price. Batch suits work that can wait for its answer.
Choosing a model for a production agent
Price is one input. Quality on your own cases, latency and the cost of retries decide the rest. Tell me what you are building.