Struct UsageReport 

pub struct UsageReport {
    pub input_tokens: u64,
    pub output_tokens: u64,
    pub total_tokens: u64,
    pub thinking_tokens: Option<u64>,
    pub cache_read_tokens: Option<u64>,
    pub cache_write_tokens: Option<u64>,
}

Uniform usage report emitted after each turn.

Normalizes provider-specific token counts into a simple, consistent structure that the platform uses for metering and billing.

Example

use adk_managed::usage::UsageReport;

let report = UsageReport::new(100, 50);
assert_eq!(report.input_tokens, 100);
assert_eq!(report.output_tokens, 50);
assert_eq!(report.total_tokens, 150);

Fields

input_tokens: u64

Number of tokens in the input/prompt.

output_tokens: u64

Number of tokens generated in the output/response.

total_tokens: u64

Total tokens (input + output). Always equals input_tokens + output_tokens.

thinking_tokens: Option<u64>

Tokens consumed by thinking/reasoning (if applicable).

cache_read_tokens: Option<u64>

Tokens read from cache (if provider supports caching).

cache_write_tokens: Option<u64>

Tokens written to cache (if provider supports caching).

Implementations

impl UsageReport

pub fn new(input_tokens: u64, output_tokens: u64) -> Self

Create a new usage report with the given input and output token counts.

Total is computed automatically as input_tokens + output_tokens.

pub fn from_usage_metadata(metadata: &UsageMetadata) -> Self

Create a UsageReport from adk-core’s UsageMetadata.

This is the primary conversion used by the session loop after each turn. It normalizes the provider-specific field names into the uniform format.

Arguments

• metadata - The raw usage metadata from the LLM response.

Example

use adk_core::UsageMetadata;
use adk_managed::usage::UsageReport;

let metadata = UsageMetadata {
    prompt_token_count: 150,
    candidates_token_count: 75,
    total_token_count: 225,
    ..Default::default()
};

let report = UsageReport::from_usage_metadata(&metadata);
assert_eq!(report.input_tokens, 150);
assert_eq!(report.output_tokens, 75);
assert_eq!(report.total_tokens, 225);

pub fn accumulate(&mut self, other: &UsageReport)

Accumulate another report into this one (for multi-turn aggregation).

This is useful for tracking total usage across an entire session.

pub fn is_empty(&self) -> bool

Returns true if this report has zero usage (no tokens consumed).

Trait Implementations

impl Clone for UsageReport

fn clone(&self) -> UsageReport

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for UsageReport

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for UsageReport

fn default() -> UsageReport

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for UsageReport

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for UsageReport

impl PartialEq for UsageReport

fn eq(&self, other: &UsageReport) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for UsageReport

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for UsageReport