Google Gemini API Pricing 2026

Google's Gemini API stands out in 2026 for two reasons: aggressive free tiers and the cheapest budget models on the market. Gemini 2.5 Flash-Lite at $0.10/M input tokens and Gemini 2.0 Flash at $0.10/M rival even DeepSeek on price.

Gemini 3.x (Preview)

Google's latest generation is in preview:

• Gemini 3.1 Pro ($2.00 input / $12.00 output) — Latest flagship, strong multimodal capabilities
• Gemini 3.1 Flash-Lite ($0.25 input / $1.50 output) — Budget preview model
• Gemini 3 Pro ($2.00 input / $12.00 output) — Stable preview option
• Gemini 3 Flash ($0.50 input / $3.00 output) — Fast and affordable

Gemini 2.5 (Production-Ready)

• Gemini 2.5 Pro ($1.25 input / $10.00 output) — Best value flagship. Cheaper than GPT-5.4 ($2.50) and Claude Opus 4.6 ($5.00).
• Gemini 2.5 Flash ($0.30 input / $2.50 output) — Excellent mid-tier option with competitive pricing
• Gemini 2.5 Flash-Lite ($0.10 input / $0.40 output) — Tied for cheapest mainstream model

Free Tier: Why Google Wins for Prototyping

Google offers generous free tiers that no other provider matches:

• Gemini Flash-Lite: 1,000 free requests/day
• Gemini 2.5 Pro: 50 free requests/day
• All Flash models have free limits for prototyping

This makes Gemini the go-to choice for prototyping and development. You can build and test your app for free before committing to paid usage.

Context Caching: 75-90% Savings

Gemini's context caching is exceptionally aggressive. Gemini 2.5 Pro cached input drops from $1.25 to $0.13/M — a 90% discount. For applications with repeated system prompts or reference documents, this can slash your bill dramatically.

How Gemini Compares

At the flagship tier, Gemini 2.5 Pro ($1.25/M input) is the cheapest premium model from a major provider — 50% less than GPT-5.4 and 75% less than Claude Opus 4.6.

At the budget tier, Gemini 2.5 Flash-Lite ($0.10/M) ties with GPT-4.1 nano for the cheapest model from a Tier 1 provider. Only DeepSeek's cached pricing ($0.028/M) is cheaper.