Struct CacheReport 

pub struct CacheReport {
    pub prefix: usize,
    pub variable: usize,
    pub turns: usize,
    pub cacheable_ratio: f64,
    pub cost_uncached: usize,
    pub cost_cached: usize,
    pub savings_ratio: f64,
}

How much a representation benefits from API prompt-caching: the per-turn prompt splits into a stable, cacheable prefix and a variable remainder. With caching the prefix is paid once at write price and thereafter at the cheap read price — so a representation with a large stable prefix is far cheaper per session.

Fields

prefix: usize

Tokens in the stable, cache-eligible prefix.

variable: usize

Tokens in the variable (non-cacheable) remainder of each turn’s prompt.

turns: usize

Session length modeled.

cacheable_ratio: f64

Fraction of the per-turn prompt that is cacheable.

cost_uncached: usize

Cost with no caching: (prefix + variable) × turns.

cost_cached: usize

Cost with prompt caching: prefix written once (×1.25), read on later turns (×0.1), plus the variable remainder every turn.

savings_ratio: f64

cost_uncached / cost_cached (≥ 1.0): how many times cheaper caching is.

Trait Implementations

impl Clone for CacheReport

fn clone(&self) -> CacheReport

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CacheReport

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

Formats the value using the given formatter.

impl Display for CacheReport

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

Formats the value using the given formatter.

impl Serialize for CacheReport

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

where __S: Serializer,

Serialize this value into the given Serde serializer.