TokenTally

Google Gemini

Gemini 3.1 Pro Preview costs

Plan 1M-token workloads, cache hits, and Search grounding fees with Google’s latest flagship preview.

Last pricing check: Mar 13, 2026

$2.00 per 1M prompt tokens$12.00 per 1M completion tokens1,000,000 token context

Why teams choose this model

Enterprise copilots with Google Search grounding
Autonomous documentation reviewers
High-accuracy analytics agents
Open main calculator

Scenario planning

Realistic cost examples

Numbers use Gemini 3.1 Pro Preview pricing

Search-grounded analyst

Analysts use Gemini 3.1 Pro + Search grounding for real-time insight summaries.

Per request

$0.021

Per month

$147.00

Tokens sent

24,500,000

2100 prompt tokens1400 completion tokens7,000 requests/mo

Policy QA bot

Gemini combs through long PDFs and outputs compliance-ready answers.

Per request

$0.0164

Per month

$246.00

Tokens sent

40,500,000

1600 prompt tokens1100 completion tokens15,000 requests/mo

Adaptive planning agent

Product teams run multi-step planning loops tied into Workspace data.

Per request

$0.0264

Per month

$158.40

Tokens sent

25,200,000

2400 prompt tokens1800 completion tokens6,000 requests/mo

Compare with

Google

Gemini 2.5 Pro Pricing Guide

Updated Mar 13, 2026

Google

Gemini 2.5 Flash Pricing

Updated Mar 13, 2026

Google

Gemini 2.5 Flash-Lite Pricing

Updated Mar 13, 2026

FAQs

How do the long-context surcharges work?

Google doubles input ($4/M) and adds a 1.5x output ($18/M) rate once prompts exceed 200K tokens. Keep scenarios under that line unless you need 1M tokens.

What about caching?

Cache reads drop to $0.20/M tokens (writes incur the full rate plus the $4.50M/hr storage fee). Use it for shared system prompts.

Is Search grounding billed separately?

Yes—5,000 prompts are free per month, then Google charges $14 per 1K Search queries in addition to the model tokens.

Pricing sources