Gemini 2.5 Flash-Lite API pricing
Gemini 2.5 Flash-Lite costs $0.10 per million input tokens and $0.40 per million output tokens at the standard rate. Here are its cached and batch prices, and what it costs for your usage.
| Token type | Standard | Batch (50% off) |
|---|---|---|
| Input | $0.10 | $0.05 |
| Cached input | $0.01 | $0.01 |
| Output | $0.40 | $0.20 |
Gemini 2.5 Flash-Lite: Audio input costs more.
- Input $3.00
- Cached input $0.3
- Output $3.60
| Workload | Standard | Batch | |
|---|---|---|---|
Support chatbot 2,000 in / 300 out tokens,
1,000 requests a day, 50% cached | $6.90 | $3.60 | 1st cheapest of 18 |
Document summaries 10,000 in / 800 out tokens,
200 requests a day | $7.92 | $3.96 | 1st cheapest of 18 |
Tagging / classifying 400 in / 20 out tokens,
20,000 requests a day | $28.80 | $14.40 | 1st cheapest of 18 |
Compare Gemini 2.5 Flash-Lite with
Other Google models
FAQ
Frequently asked questions
How much does Gemini 2.5 Flash-Lite cost per million tokens?
Gemini 2.5 Flash-Lite costs $0.10 per million input tokens and $0.40 per million output tokens at the standard rate. Prices are in USD, before tax, from Gemini API pricing, last checked on 30 Sept 2026. Audio input costs more.
Does Gemini 2.5 Flash-Lite have a Batch API discount?
Yes. Through the Batch API, Gemini 2.5 Flash-Lite costs $0.05 per million input tokens and $0.20 per million output tokens, 50% less than the standard rate. Batch requests are processed asynchronously rather than straight away.
How much does prompt caching save on Gemini 2.5 Flash-Lite?
Input read from the prompt cache costs $0.01 per million tokens instead of $0.10, 90% less. Writing to or storing the cache can cost extra; those charges aren’t included here.
Is Gemini 2.5 Flash-Lite one of the cheaper API models?
For a support chatbot sending 1,000 requests a day (2,000 input and 300 output tokens each, 50% of the input cached), Gemini 2.5 Flash-Lite is the 1st cheapest of the 18 OpenAI, Anthropic and Google models we track, at $6.90 a month. Price alone doesn’t tell you how well a model will handle your task, so test the candidates on your own prompts.