Google DeepMind · Updated 2026-07-22

Google Gemini API Pricing 2026: Gemini 2.5 Pro & Flash Costs

Per-million-token pricing for the current Gemini API lineup — 2.5 Pro, 2.5 Flash and 2.0 Flash. Every model ships a 1 million-token context window, and Gemini 2.0 Flash is among the cheapest frontier APIs on the market. Cross-compare on the main calculator.

1M token context window

All three Gemini models support a 1,000,000-token context window — roughly 5× larger than OpenAI o3/o4-mini (200K) and ~8× GPT-4o (128K). Pass entire codebases, long PDFs or multi-hour transcripts in a single call without chunking.

Cheapest Gemini model

Gemini 2.0 Flash $0.10 input / $0.40 output per million tokens. One of the lowest per-million-token rates of any frontier-class API.

Google Gemini pricing per million tokens

USD per 1,000,000 tokens. Google does not publish separate batch rates in this dataset; cached rate applies to repeated context.

ModelInput $/MOutput $/MCached $/MContext
Gemini 2.0 Flash$0.10$0.40$0.0251M
Gemini 2.5 Flash$0.15$0.60$0.03751M
Gemini 2.5 Pro$1.25$10.00$0.31251M

Gemini 2.0 Flash

$0.10 in / $0.40 out · 1M context

ChatCodeVision

Cache saves 75% on input

Gemini 2.5 Flash

$0.15 in / $0.60 out · 1M context

ChatCodeVision

Cache saves 75% on input

Gemini 2.5 Pro

$1.25 in / $10.00 out · 1M context

ChatCodeVisionReasoning

Cache saves 75% on input

When Gemini wins on cost

  • Very long context: 1M-token window avoids the chunking and embedding cost that smaller-context models incur.
  • Bulk multimodal: Gemini Flash pairs cheap per-token rates with native vision support.
  • Input-heavy workloads: Gemini 2.5 Pro input ($1.25/M) undercuts GPT-4o ($2.50) and Claude Sonnet 4 ($3.00).

Compare Gemini against other providers

Frequently asked questions

How much does Gemini 2.5 Pro cost?

Gemini 2.5 Pro lists at $1.25 per million input tokens and $10.00 per million output tokens. Prompt caching cuts repeated input to $0.3125/M — a 75% discount. Gemini 2.5 Pro is notably cheaper on input than GPT-4o ($2.50) and Claude Sonnet 4 ($3.00), while carrying a 1 million-token context window.

Is Gemini Flash free?

Google offers a limited free tier on Gemini Flash for development and low-volume use (subject to per-minute rate limits), but production API calls are billed. Gemini 2.0 Flash is the cheapest paid option in the lineup at $0.10/M input and $0.40/M output — among the lowest per-million-token rates of any frontier provider.

What is the Gemini 1M token context window worth?

Every current Gemini model supports a 1,000,000-token context window — roughly 5× OpenAI's 200K (o3/o4-mini) and 8× GPT-4o's 128K. This matters for workloads that pass entire codebases, long documents or multi-hour transcripts in a single call. The cost trade-off is input volume: a full 1M-token prompt on Gemini 2.5 Pro costs $1.25 per call in input alone.

Gemini 2.5 Flash vs 2.0 Flash — which should I use?

Gemini 2.5 Flash ($0.15/$0.60) is the newer generation with stronger reasoning and multimodal performance, while 2.0 Flash ($0.10/$0.40) is cheaper. For new production workloads, 2.5 Flash is usually worth the small premium; 2.0 Flash remains a strong pick for bulk classification and high-volume chat where every cent per million tokens counts.

Does Gemini support a batch API?

Google offers the Gemini Batch API for asynchronous workloads with similar 50% discount economics to OpenAI and Anthropic. Batch is suited to offline processing where 24-hour turnaround is acceptable. The rates shown in the table below are standard on-demand prices.