Google Gemini API Pricing (February 2026)
Current pricing for all Gemini models. Google offers extremely competitive pricing, especially for Flash models. Prices shown are per 1 million tokens.
| Model | Input Price | Output Price | Context Window |
|---|---|---|---|
| Gemini 3 ProFlagship | $1.25 / 1M | $5.00 / 1M | 2M tokens |
| Gemini 3 FlashBest Value | $0.075 / 1M | $0.30 / 1M | 2M tokens |
| Gemini 2.5 Pro | $1.25 / 1M | $5.00 / 1M | 1M tokens |
| Gemini 2.5 Flash | $0.075 / 1M | $0.30 / 1M | 1M tokens |
Tokens context window, the largest of any major AI model
Equivalent to 1.5 million words or 3,000+ pages
Free Tier Available
Google offers a generous free tier through AI Studio:
- Gemini 3 Flash: 15 requests/minute, 1,500 requests/day
- Gemini 3 Pro: 2 requests/minute, 50 requests/day
- Perfect for prototyping and small projects
Why Gemini 3 Flash is So Cheap
Gemini 3 Flash at $0.075/1M tokens is one of the cheapest production-ready AI models available. Here's how it compares:
| Model | Input Price | Price Ratio |
|---|---|---|
| Gemini 3 Flash | $0.075 / 1M | 1x (baseline) |
| DeepSeek V3 | $0.07 / 1M | 0.9x (slightly cheaper) |
| GPT-4o-mini | $0.15 / 1M | 2x more expensive |
| Claude Haiku 4.5 | $0.25 / 1M | 3.3x more expensive |
| GPT-4o | $2.50 / 1M | 33x more expensive |
Pro Tip: Context Caching
Gemini offers context caching for 75% discount on cached tokens. Create a cached context using the API for repeated prompts. Storage costs $0.001/1K tokens/hour with minimum 5-minute TTL.
When to Use Gemini
Gemini 3 Pro. Best For:
- Complex reasoning and analysis tasks
- Long document processing (2M context)
- Code generation and review
- Multimodal tasks (image, video, audio)
Gemini 3 Flash. Best For:
- High-volume, cost-sensitive workloads
- Simple classification and extraction
- Real-time applications requiring speed
- Prototyping and development
Frequently Asked Questions
What is the cheapest AI model in 2026?
Gemini 3 Flash at $0.075/1M tokens and DeepSeek V3 at $0.07/1M are the cheapest production-ready AI models in 2026. Both offer excellent quality for their price point.
Does Gemini have a free tier?
Yes! Google AI Studio offers a free tier. Gemini 3 Flash: 15 requests/minute, 1,500/day. Gemini 3 Pro: 2 requests/minute, 50/day. Perfect for prototyping.
How big is Gemini's context window?
Gemini 3 models support up to 2 million tokens, the largest of any major AI model. This is approximately 1.5 million words or 3,000+ pages of text.
Gemini vs GPT-4o, which is better value?
For cost, Gemini wins. Gemini 3 Flash is 33x cheaper than GPT-4o. Gemini 3 Pro is 2x cheaper. GPT-4o may offer slightly better quality for some tasks, but Gemini provides excellent value.