Google Gemini API Pricing Calculator
Estimate your Gemini API token costs across Gemini 2.5 Pro, Gemini 2.0 Flash, Flash Lite, and more. Calculate per-request, daily, and monthly spend — including context caching savings.
Estimate Your Gemini API Costs
Select a model, enter your usage, and get an instant cost breakdown
Prices are per 1 million tokens (input / output). Updated June 2026. Gemini 1.5 Flash and Flash-8B have a free tier via Google AI Studio.
System prompt + user message + any document context. ≈ 600 words per 800 tokens.
The model’s generated response. ≈ 300 words per 400 tokens. Output tokens cost more than input.
Tokens already stored in cache (e.g., system prompt, document). Cached tokens are billed at ~25% of the normal input rate — a 75% saving.
Total API calls across all users or automations per day.
Use 30 for always-on services, or fewer for weekday-only tools.
Cost Estimate
Google Gemini API pricing breakdown
Select a Gemini model and enter your usage details, then click Calculate to see your estimated API costs.
Gemini API Pricing Guide 2026
Current Google Gemini API prices per model. All costs in USD per 1 million tokens. Context caching rates shown where available. Verify the latest pricing at ai.google.dev/pricing.
| Model | Context | Input ($/1M) | Output ($/1M) | Cached Input | Notes |
|---|---|---|---|---|---|
| Gemini 2.5 Pro | 1M tokens | $1.25 | $10.00 | $0.31 | Thinking New |
| Gemini 2.5 Flash | 1M tokens | $0.30 | $2.50 | $0.075 | Thinking |
| Gemini 2.5 Flash-Lite | 1M tokens | $0.10 | $0.40 | $0.025 | Fast |
| Gemini 2.0 Flash | 1M tokens | $0.075 | $0.30 | $0.01875 | Fast 1M ctx |
| Gemini 2.0 Flash-Lite | 1M tokens | $0.0375 | $0.15 | — | Fast |
| Gemini 1.5 Pro | 2M tokens | $1.25 | $5.00 | $0.3125 | 2M ctx |
| Gemini 1.5 Flash | 1M tokens | $0.075 | $0.30 | $0.01875 | Free tier |
| Gemini 1.5 Flash-8B | 1M tokens | $0.0375 | $0.15 | $0.01 | Free tier |
| text-embedding-004 | 2K tokens | $0.013 | — | — | Embeddings only |
Gemini API Pricing FAQ
Everything you need to know about Google Gemini API costs, tokens, context caching, and billing.
Google Gemini API pricing is based on token consumption, billed separately for input tokens (your prompts, system instructions, and context) and output tokens (the model’s generated responses). Costs are expressed in USD per 1 million tokens ($/1M tokens). Some models like Gemini 1.5 Flash have a free tier with generous rate limits, making them ideal for development and low-volume production workloads.
Gemini 2.5 Pro is Google’s most capable and intelligent model, optimised for complex reasoning, coding, and long-context tasks. It includes a built-in thinking mode and supports up to 1 million token context. Gemini 2.0 Flash is a faster, significantly cheaper model designed for high-throughput applications where speed and cost matter more than maximum capability. For most production applications, Flash offers the best cost-to-performance ratio.
Yes. Google offers a free tier for Gemini API via Google AI Studio with rate limits of 15 requests per minute (RPM) for Gemini 1.5 Flash and lower limits for Gemini 2.5 Pro. The free tier is suitable for experimentation, prototyping, and low-volume personal projects. For production workloads, the paid tier removes rate limits and provides SLAs.
Context caching allows you to store frequently reused content (like long system prompts, documents, or conversation history) once and reference it cheaply in subsequent requests. Cached input tokens are billed at roughly 25% of the standard input rate — a 75% saving. This is particularly valuable for RAG (retrieval-augmented generation) systems, document Q&A, or applications with large, repeated system prompts.
In Gemini models, a token is approximately 4 characters or 0.75 words of English text. For images, each image counts as approximately 258 tokens at standard resolution. Audio is counted at 32 tokens per second. Video frames are counted individually. Google provides a token counting endpoint in the API so you can measure exact token usage before making full requests.
Gemini models are generally among the most affordable frontier models available. Gemini 2.0 Flash at $0.075/$0.30 per 1M tokens is significantly cheaper than GPT-4o ($2.50/$10.00) and Claude Sonnet ($3.00/$15.00). Gemini 2.5 Pro at $1.25/$10.00 per 1M tokens is competitive with GPT-4o for complex tasks while offering a much larger 1M token context window. For cost-sensitive applications, Gemini 2.0 Flash-Lite is among the cheapest capable models available.
Gemini models offer some of the largest context windows of any commercial LLM. Gemini 2.5 Pro and 2.0 Flash support up to 1 million tokens of context (approximately 700,000 words or 1,500 pages of text). Gemini 1.5 Pro supports up to 2 million tokens. This makes Gemini particularly well-suited for tasks involving entire codebases, long legal documents, or book-length content.
The Gemini API is accessible through two main platforms: Google AI Studio (ai.google.dev), which is the quickest way to get started with an API key and includes the free tier; and Google Cloud Vertex AI, which is recommended for enterprise production workloads and offers additional features like VPC Service Controls, regional endpoints, and enterprise SLAs.