If you're building with Google's Gemini API, you've probably asked yourself:
"How much is this actually going to cost me per month?"
The official pricing page gives you per-token rates, but calculating real-world costs — with different models, batch vs standard pricing, and monthly volume — is a headache. So I built a free tool to solve it.
What It Does
- All Gemini models — 2.5 Flash, 3.5 Flash, 2.5 Pro, Flash-Lite, and legacy models
- 3 pricing tiers — Standard, Batch (50% cheaper), and Priority
- Cross-provider comparison — Stack Gemini against GPT-4o, Claude, and more
- Monthly cost projection — Enter your expected volume, get a dollar amount instantly
- Free tier info — See which models offer free daily quotas
Quick Example
Running 100K requests/month with 1K input + 500 output tokens per request:
| Model | Monthly Cost |
|---|---|
| Gemini 2.5 Flash-Lite | ~$4.50 |
| Gemini 2.5 Flash | ~$28 |
| GPT-4o Mini | ~$22 |
| GPT-4o | ~$225 |
The cheapest Gemini model costs less than a coffee per month at moderate scale.
Why I Built It
I've been building AI tools and the first question every time is "what's the bill going to look like?" No calculator I found supported all Gemini models with batch pricing. So I built one.
It's 100% free, no signup, no tracking. Just a tool I wanted to exist.
Feedback welcome! What features would you add?
Top comments (0)