Gemini 2.5 Flash pricing
Google Gemini charges $0.30 per million input tokens and $2.50 per million output tokens for Gemini 2.5 Flash, so output costs 8.3 times input. Cached input is $0.03, 90% off the input rate. The Batch API takes 50% off. We first recorded this model on 3 Oct 2026 and have 4 rows for it since.
Current price
US dollars per million tokens| Rate | Price per 1M tokens |
|---|---|
| Input | $0.30 |
| Output | $2.50 |
| Cached input | $0.03 |
| Cache write (5 minute) | $0.03 |
| Batch input | $0.15 |
| Batch output | $1.25 |
Source: Google Gemini's pricing page, price since (effective date stated by the page). Rates the page does not print for this model are not listed.
What Gemini 2.5 Flash costs you
Our calculation from the prices above, at three usage levels. Same levels on every model page, so they compare directly. The cache column assumes half of input is a cache hit. Batch assumes every request goes through the Batch API. Before tax, no volume discount.
| Usage | Tokens a month (in / out) | Standard | A year, standard | 50% cached input | Batch |
|---|---|---|---|---|---|
| Lighta chatbot or script, about 1,000 requests a month | 2M / 500K | $1.85 | $22.20 | $1.58 | $0.93 |
| Mediuma product feature, about 20,000 requests a month | 40M / 10M | $37.00 | $444.00 | $31.60 | $18.50 |
| Heavyan agent or pipeline at volume, about 200,000 requests a month | 400M / 100M | $370.00 | $4,440.00 | $316.00 | $185.00 |
Try your own numbers in the cost calculator, opened on Gemini 2.5 Flash and the alternatives below.
Price history
The input and output price of Gemini 2.5 Flash has not changed since we first read it on 3 Oct 2026; the rows below record other rates first shown or the model leaving the page.
| Date | Input | Output | Cached input | What changed |
|---|---|---|---|---|
| $0.30 | $2.50 | $0.03 | First recorded | |
| $0.30 | $2.50 | $0.03 | Batch input first shown, $0.15; Batch output first shown, $1.25 | |
| $0.30 | $2.50 | $0.03 | Values reread, no price change | |
| $0.30 | $2.50 | $0.03 | Cache write first shown, $0.03 |
A row is recorded only when a value on Google Gemini's page differs from the last one we read, so the dates are when we saw each change.
Gemini 2.5 Flash alternatives by price
The three current models whose price is nearest to Gemini 2.5 Flash's, blending input and output 3 to 1 (our calculation from the current table, one model per provider).
| Model | Input | Output | Cached input | Medium usage a month | Compared with Gemini 2.5 Flash |
|---|---|---|---|---|---|
| Google Gemini Gemini 3.5 Flash-Lite | $0.30 | $2.50 | $0.03 | $37.00 | Same |
| Qwen qwen3.6-35b-a3b | $0.375 | $2.25 | – | $37.50 | $0.50 more |
| Mistral Mistral Large 4 | $0.70 | $2.10 | $0.07 | $49.00 | $12.00 more |
MethodWhere these prices come from
Prices come from our LLM API pricing timeline, read once a day from each provider's own pricing page; a value is kept only if the same number appears in the page text. – means the page does not publish that price. The monthly costs and alternatives are our arithmetic on those prices, not figures from Google Gemini.