API Billing Guide
Kimi API is billed based on token usage. Prices vary by model and feature.
Billing Basics
- Billed by token: Each API call is charged separately based on input tokens and output tokens.
- token unit: 1M = 1,000,000 tokens.
- Prices vary by model: More capable models have higher unit prices. Choose the right model for your business needs.
Add-on Feature Billing
- Web search: Each call that uses web search is charged an additional ¥0.03, separate from token usage.
Context Caching
Context Caching lets you cache frequently used context, such as system prompts and reference documents. tokens served from a cache hit are billed at a discounted rate, helping reduce the cost of repeated context.
For detailed Context Caching pricing, see the official documentation.
Billing Details
For the complete model pricing table and billing rules, visit:
https://klmi.io/docs/pricing/chat