Our billing system tracks your usage in real-time and charges you based on the number of tokens processed by each model. Different models have different pricing tiers based on their capabilities and computational requirements.

For more detailed information about billing, subscriptions, and credits, please visit our billing documentation.

Tokens are the fundamental units of text processing in AI models. They represent pieces of text that can be:

• Words or parts of words
• Punctuation marks
• Spaces and special characters

Quick estimation: On average, 1 token ≈ 4 characters or ≈ 0.75 words in English. A typical sentence contains 15-20 tokens.

To understand how tokens work in real time, you can use the OpenAI Tokenizer tool to see exactly how your text is broken down into tokens.

Our pricing distinguishes between two types of tokens:

• Input Tokens: Text you send to the model (prompts, context, system messages)
• Output Tokens: Text generated by the model in response

Why different pricing? Output tokens typically cost more because:

• Generation requires more computational resources
• Each output token depends on processing all previous tokens
• Models must maintain context throughout generation

Our pricing is based on token usage, with different models having different pricing tiers based on their capabilities and computational requirements. You can view detailed pricing information on our pricing page.

We generally do not provide refunds for purchased content, except in cases where token usage is below $2 credits or equivalent, or as required by mandatory legislation. For complete details, please read our Terms of Service.