TokenTally

Google DeepMind

Gemini 1.5 Pro token costs

Feed entire transcripts or video frames into Gemini’s 2M-token context. These scenarios show what that costs in practice.

Last pricing check: Mar 13, 2026

$3.50 per 1M prompt tokens$10.50 per 1M completion tokens2,000,000 token context

Why teams choose this model

Video + meeting understanding
Large-document QA
Multi-modal research copilots
Open main calculator

Scenario planning

Realistic cost examples

Numbers use Gemini 1.5 Pro pricing

Meeting IQ

Full video transcripts summarized with action items + owners.

Per request

$0.0193

Per month

$34.65

Tokens sent

5,940,000

2200 prompt tokens1100 completion tokens1,800 requests/mo

Audit digest

Teams upload 200+ page docs and get clause-level highlights.

Per request

$0.0186

Per month

$22.26

Tokens sent

4,200,000

2600 prompt tokens900 completion tokens1,200 requests/mo

Media research

Analysts query mixed text/images for insights.

Per request

$0.0136

Per month

$47.78

Tokens sent

8,750,000

1800 prompt tokens700 completion tokens3,500 requests/mo

Compare with

Google

Gemini 1.5 Flash Cost Calculator

Updated Mar 13, 2026

Google

Gemini 2.5 Pro Pricing Guide

Updated Mar 13, 2026

Google

Gemini 3.1 Pro Preview Pricing

Updated Mar 13, 2026

FAQs

How does Gemini bill audio/video inputs?

Uploads are tokenized the same way text is—TokenTally’s counters still apply once the content is transcribed internally.

Do I need Vertex AI for billing?

Production teams usually deploy via Vertex for governance, but pricing per token stays identical whether you call Gemini API or Vertex endpoints.

How do I cap runaway prompts?

Set per-request limits server-side. TokenTally’s shareable scenarios help you document those caps for engineering + finance.

Pricing sources