Number of tokens that were served from cache.
Indicates how many input tokens were reused from previous processing through caching mechanisms, reducing computational costs and improving response times. Higher cached token usage indicates more efficient resource utilization and better system optimization.
Total amount of input tokens consumed.
Represents the complete input token usage including both newly processed tokens and cached tokens that were reused from previous operations. This total provides the baseline for input processing cost analysis.
Detailed input token usage statistics with caching analysis.
Provides comprehensive tracking of input token consumption, distinguishing between fresh token processing and cached token reuse. This analysis is crucial for understanding the efficiency of context caching and optimizing input processing strategies to reduce computational costs.