# Gemini vs DeepSeek: API prices compared, dated and sourced

Canonical page: https://bertrande.com/en/vs/gemini-vs-deepseek/
License: CC-BY-4.0, reuse with attribution and a link.

No value below was produced, estimated or completed by a language model. Every figure was read from a source we name, on the date we give.

The cheapest Gemini model we track, Gemini 2.5 Flash-Lite, costs $0.400 per million output tokens and an input price that depends on the type of content. The cheapest DeepSeek model, deepseek-flash, costs $1.20 per million output tokens and $0.300 per million input tokens. At the entry level, Gemini is 3.0 times cheaper than DeepSeek on output tokens. Prices read on 11 September 2026 and 18 September 2026.

## Gemini models

| Model | Input $/1M | Output $/1M | Read on |
|---|---|---|---|
| Gemini 2.5 Flash-Lite | by content type | $0.400 | 2026-09-18 |
| Gemini 3.1 Flash-Lite | by content type | $1.50 | 2026-09-18 |
| Gemini 2.5 Flash | by content type | $2.50 | 2026-09-18 |
| Gemini 3.5 Flash-Lite | by content type | $2.50 | 2026-09-18 |
| Gemini 3 Flash Preview | by content type | $3.00 | 2026-09-18 |
| Gemini 3.6 Flash | $0.750 | $3.75 | 2026-09-18 |
| Gemini 3.7 Flash | $0.750 | $3.75 | 2026-09-18 |
| Gemini 3.8 Flash | $0.750 | $3.75 | 2026-09-18 |
| Gemini Robotics ER 1.6 Preview | by content type | $5.00 | 2026-09-11 |
| Gemini Robotics ER 2 Preview | by content type | $5.00 | 2026-09-17 |
| Gemini Robotics ER 2 Streaming Preview | by content type | $5.00 | 2026-09-17 |
| Gemini 3.5 Flash | $1.50 | $9.00 | 2026-09-18 |

## DeepSeek models

| Model | Input $/1M | Output $/1M | Read on |
|---|---|---|---|
| deepseek-flash | $0.300 | $1.20 | 2026-09-11 |
| deepseek-v4-pro | $1.32 | $3.96 | 2026-09-11 |


Method: https://bertrande.com/en/methodology/
Report an error: https://bertrande.com/en/corrections/