Gemini API Cost Calculator

About This Tool

The Gemini API Cost Calculator handles all three Gemini 2.5 tiers — Pro, Flash, and Flash-Lite — including the long-context pricing kink that hits Pro at 200K input tokens. Enter token counts and request volume, and the tool computes per-call and monthly cost across the lineup.

Gemini pricing (April 2026)

Gemini 2.5 Pro: $1.25 input / $5 output per million tokens up to 200K input tokens. Beyond 200K, the rate doubles to $2.50 / $10 to reflect the cost of attending over very long contexts. Gemini 2.5 Flash: $0.30 / $2.50 with a 1M-token context window. Flash-Lite: $0.10 / $0.40 — the cheapest production-grade model from any major lab.

The 1M-token context window

Pro's defining feature is its 1 million token context. That fits roughly 1500 pages of text or 50 hours of audio transcript. For document Q&A and codebase analysis, this lets you skip RAG entirely and stuff the full corpus into the prompt — at the cost of higher per-call spend and slower latency. The economics break even with RAG around 50K-100K total tokens; below that, stuffing wins.

When to pick which Gemini

Pro for hard reasoning, long documents, and multimodal work. Flash for everyday chat and structured extraction at scale — it's often the best price-per-performance option in the market. Flash-Lite for ultra-high-volume classification, routing, and lightweight agents where the cheapest token is the right token.

Compare against alternatives: GPT API cost calculator, Claude API cost calculator, and the LLM cost comparison. For long-context budgeting, see the context window calculator. Estimate token counts with the token counter.