Gemini Cost Calculator
Estimate Google Gemini API token spend with cached-input, Batch, and long-context pricing.
Inputs
Choose the model tier you plan to use in production.
Average prompt size per API call — system prompt + context + user message. Typical ranges: simple chat 500–2,000 · RAG chatbot 2,000–8,000 · document analysis 10,000–100,000. A page of text ≈ 750 tokens.
Average response length per API call. Typical ranges: short answer 200–500 · paragraph 500–1,000 · long-form 1,000–4,000 tokens.
Total API calls per month across all users and jobs. 10 users × 3 sessions/day × 10 calls/session × 30 days = 9,000 requests/month.
Fraction of input tokens billed as cached reads. Cache writes and storage are not included. Use 0 if caching is disabled.
((2,000 fresh input × $1.25 + 0 cached-read input × $0.125 + 500 output × $10) / 1M × 10,000 req) × 12
Decision Summary
Enabling Prompt Cache with a 60% hit ratio could save $162.00/yr (18% reduction). Fix your system prompt at the start of every request and reuse it across calls.
Switching to Gemini 2.5 Flash cuts estimated cost by 75% — saving $678.00/yr. Run an A/B test against a representative eval set before changing production traffic.
Cost breakdown
| Item | Monthly | Yearly |
|---|---|---|
| Fresh input (2,000 tok/req × 10,000 req × $1.25/MTok) | $25.00 | $300.00 |
| Output spend (500 tok/req × 10,000 req × $10/MTok) | $50.00 | $600.00 |
| Total monthly spend | $75.00 | $900.00 |
Comparison
| Option | Monthly | Yearly |
|---|---|---|
| OpenAI GPT-5 | $75.00 | $900.00 |
| Claude Claude Sonnet 5 | $90.00 | $1,080.00 |
| Gemini Gemini 2.5 Procurrent | $75.00 | $900.00 |
| DeepSeek DeepSeek V4 Flashcheapest | $4.20 | $50.40 |
Pricing sources
Last verified 2026-08-03 · Gemini official pricing ai.google.dev/gemini-api/docs/pricing
Industry Benchmark
Pricing sources
Last verified 2026-08-03 · developers.openai.com/api/docs/models/compare developers.openai.com/api/docs/models/compare · platform.claude.com/docs/en/about-claude/pricing platform.claude.com/docs/en/about-claude/pricing · ai.google.dev/gemini-api/docs/pricing ai.google.dev/gemini-api/docs/pricing · api-docs.deepseek.com/quick_start/pricing api-docs.deepseek.com/quick_start/pricing
Trends & comparison
Trend
Comparison (monthly vs. yearly)
Official Google rates used
Verified August 3, 2026: Gemini 2.5 Pro is $1.25/MTok input and $10 output up to 200,000 input tokens, then $2.50/$15. Gemini 3.6 Flash is $1.50/$7.50, Gemini 2.5 Flash is $0.30/$2.50, and Flash-Lite is $0.10/$0.40.
A reproducible workload example
At 10,000 monthly calls with 1,000 input and 500 output tokens, Gemini 2.5 Flash costs $15.50 and Gemini 2.5 Pro costs $62.50 before cache, Batch, storage, tools, or other service charges.
Context caching scope
The calculator uses the current cached-input read rates in the official Google table: $0.125/MTok for Gemini 2.5 Pro below the long-context threshold, $0.15 for Gemini 3.6 Flash, $0.03 for Gemini 2.5 Flash, and $0.01 for Flash-Lite. Cache storage is not included.
Price is only one routing constraint
This page does not infer model quality from provider labels or benchmark summaries. Compare models on your own eval set and include latency, rate limits, context use, data policy, tool pricing, and regional availability in the final decision.
Frequently asked questions
What Gemini token rates does this calculator use?▾
Gemini 2.5 Pro starts at $1.25/MTok input and $10 output; Gemini 3.6 Flash is $1.50/$7.50; Gemini 2.5 Flash is $0.30/$2.50; and Flash-Lite is $0.10/$0.40.
When does Gemini 2.5 Pro long-context pricing apply?▾
Prompts above 200,000 input tokens use $2.50/MTok input, $15 output, and $0.25 cached-input pricing for the full request. The calculator applies this tier automatically.
Does the cache estimate include storage?▾
No. The cache slider models cached-input reads only. Explicit cache storage and other provider charges are excluded because they depend on model, retained token volume, and duration.
Is Flash or Pro more cost-effective?▾
Flash has lower published token prices, but cost-effectiveness also depends on whether it passes your quality, latency, context, and feature requirements. Test the same production workload before choosing.
Related calculators
Compare verified first-party API prices on the same standard workload → AI Model Price Leaderboard