NeuralPlus 0.2.0
A small C++17 SDK for AI clients, sessions, tools, and tracing
Loading...
Searching...
No Matches
neuralplus::TokenUsage Struct Reference

Normalized token counts. Providers may leave unavailable values empty. More...

#include <types.hpp>

Public Attributes

std::optional< std::size_t > input_tokens
 Total tokens consumed from input, including provider cache reads/writes.
 
std::optional< std::size_t > output_tokens
 Tokens generated as output.
 
std::optional< std::size_t > total_tokens
 Provider-reported or safely calculated total.
 
std::optional< std::size_t > cached_input_tokens
 Cached portion of input tokens, when reported.
 
std::optional< std::size_t > cache_creation_input_tokens
 Input tokens written to a provider cache, when reported separately.
 
std::optional< std::size_t > reasoning_tokens
 Reasoning/thinking tokens, when reported.
 

Detailed Description

Normalized token counts. Providers may leave unavailable values empty.


The documentation for this struct was generated from the following file: