Updated February 2026
AI API Pricing Calculator
Calculate and compare costs for ChatGPT, Claude, Gemini, Llama, and 25+ AI models. Get accurate monthly estimates based on your actual usage patterns.
25+
AI Models
8
Providers
Real-time
Pricing
Configure Usage
Select Provider
Select Model
Monthly Usage
Quick Presets
šØ Hobbyist
š Startup
š¼ Business
š¢ Enterprise
Complete AI API Pricing Reference 2026
All prices per 1 million tokens. 1M tokens ≈ 750,000 words.
| Provider | Model | Input | Output | Context | Best For |
|---|---|---|---|---|---|
| OpenAI | GPT-4oPopular | $2.50 | $10.00 | 128K | General, vision |
| OpenAI | GPT-4o MiniCheap | $0.15 | $0.60 | 128K | Cost-effective |
| OpenAI | o1 | $15.00 | $60.00 | 200K | Complex reasoning |
| Anthropic | Claude 3.5 SonnetBest Code | $3.00 | $15.00 | 200K | Coding, writing |
| Anthropic | Claude 3 HaikuFast | $0.25 | $1.25 | 200K | Quick tasks |
| Gemini 1.5 Pro | $1.25 | $5.00 | 2M | Long context | |
| Gemini 1.5 FlashCheapest | $0.075 | $0.30 | 1M | High volume | |
| Meta | Llama 3.1 70B | $0.35 | $0.40 | 128K | Open source |
| Mistral | Mistral Large | $2.00 | $6.00 | 128K | Multilingual |
| Groq | Llama 3.1 70BUltra Fast | $0.59 | $0.79 | 128K | Speed critical |
Cost Optimization Tips
Start Small
Use GPT-4o Mini or Gemini Flash for simple tasks. Only upgrade to premium models when quality matters.
Optimize Prompts
Shorter, clearer prompts use fewer tokens. Remove filler words and unnecessary context.
Cache Responses
Store frequent queries locally. Avoid repeated API calls for identical requests.
Frequently Asked Questions
What is a token in AI pricing?
A token is roughly 4 characters or 0.75 words in English. "Hello world" is about 2-3 tokens. AI APIs charge separately for input (your prompt) and output (AI response) tokens.
Which AI model is cheapest?
Google Gemini 1.5 Flash ($0.075/1M input) is currently the cheapest major model. GPT-4o Mini ($0.15/1M) and Claude 3 Haiku ($0.25/1M) are also very affordable options.
Why is output more expensive than input?
Generating output requires more computation than processing input. The model must predict each token sequentially, using more GPU resources than reading your prompt.
Is there free AI API access?
Yes! Google offers generous free tiers for Gemini. OpenAI provides $5-18 free credits for new accounts. Groq, Together, and others have free tiers with rate limits.