Nova 2 Lite vs Gemini 3.5 Flash-Lite
Amazon against Google, costed on the same three workloads. Neither is cheaper everywhere — which one wins depends on how much you read versus write.
Run this comparison on your own numbers
Monthly bill, side by side
| Workload | Nova 2 Lite | Gemini 3.5 Flash-Lite | Difference |
|---|---|---|---|
| Support chatbot 2,500 six-turn conversations a day |
$1,613 | $1,613 | the same |
| Document summarizer 1,000 single-shot summaries a day |
$170 | $170 | the same |
| Coding agent 200 twelve-turn sessions a day |
$1,022 | $1,022 | the same |
No prompt caching and no batch discount on either side, so the comparison is like for like. Both figures assume 30 days a month.
The rate cards
| Nova 2 Lite | Gemini 3.5 Flash-Lite | |
|---|---|---|
| Input, per 1M tokens | $0.3 | $0.3 |
| Output, per 1M tokens | $2.5 | $2.5 |
| Cached input | $0.075 | $0.03 |
| Context window | 1M | 1.05M |
| Maximum output | 64K | 66K |
| Reasoning model | yes | yes |
| Vision | yes | yes |
Beyond the price
- Gemini 3.5 Flash-Lite reads up to 1.05M tokens of context in one request; Nova 2 Lite stops at 1M.
- Gemini 3.5 Flash-Lite can write up to 66K tokens in one response, against 64K for Nova 2 Lite.
- Cached input costs 75% less than fresh input on Nova 2 Lite and 90% less on Gemini 3.5 Flash-Lite, which matters most for workloads that resend the same prompt.
Run this comparison on your own numbers
Full detail: Nova 2 Lite pricing · Gemini 3.5 Flash-Lite pricing
More comparisons
Other head-to-heads that include Nova 2 Lite or Gemini 3.5 Flash-Lite.
- Nova 2 Lite vs Claude Haiku 4.5
- Nova 2 Lite vs Qwen3 Next 80B
- Nova 2 Lite vs DeepSeek V4 Flash
- Gemini 3.5 Flash-Lite vs Claude Haiku 4.5
- Gemini 3.5 Flash-Lite vs Qwen3 Next 80B
- Gemini 3.5 Flash-Lite vs DeepSeek V4 Flash
Common questions
Which is cheaper, Nova 2 Lite or Gemini 3.5 Flash-Lite?
Neither is cheaper everywhere — which one wins depends on how much you read versus write.
How much would Nova 2 Lite and Gemini 3.5 Flash-Lite cost each month?
On PromptSpend's standard workloads, per month: support chatbot: Nova 2 Lite $1,613, Gemini 3.5 Flash-Lite $1,613 (the same on both); document summarizer: Nova 2 Lite $170, Gemini 3.5 Flash-Lite $170 (the same on both); coding agent: Nova 2 Lite $1,022, Gemini 3.5 Flash-Lite $1,022 (the same on both). No prompt caching and no batch discount is assumed.
How do Nova 2 Lite and Gemini 3.5 Flash-Lite differ apart from price?
Gemini 3.5 Flash-Lite reads up to 1.05M tokens of context in one request; Nova 2 Lite stops at 1M. Gemini 3.5 Flash-Lite can write up to 66K tokens in one response, against 64K for Nova 2 Lite. Cached input costs 75% less than fresh input on Nova 2 Lite and 90% less on Gemini 3.5 Flash-Lite, which matters most for workloads that resend the same prompt.