Tarification et Coûts
Input Price
The per-million-token cost charged for tokens in your prompt.
Input price (also called prompt price) covers all tokens the model reads: system prompt, user message, conversation history, and retrieved context. Input tokens are processed during the 'prefill' phase, which is faster and cheaper than generation. Input prices typically range from $0.05 to $15 per million tokens depending on model tier.
Termes Associés
Output Price
The per-million-token cost charged for tokens the model generates.
Token
The basic unit of text that language models process and are billed by.
Context Caching
A provider feature that stores a reusable prefix in memory to avoid re-processing repeated tokens.
Batch Processing
Submitting requests asynchronously in bulk for a 50% price discount.