Struct PrefillCost 

pub struct PrefillCost {
    pub new_blocks: usize,
    pub new_tokens: usize,
    pub cached_tokens: usize,
    pub active_cached_tokens: usize,
}

Represents the cost of prefilling content in the cache

Fields

new_blocks: usize

new_tokens: usize

cached_tokens: usize

Number of tokens already cached (prefix hit). isl = cached_tokens + new_tokens

active_cached_tokens: usize

Subset of cached_tokens backed by active blocks. Physical-capacity admission discounts only these because inactive reuse is re-consumed.

Implementations

impl PrefillCost

pub fn predict_prefill_compute( &self, new_tokens: Option<usize>, perf_model: &PerfModel, ) -> Result<f64>

Trait Implementations

impl Clone for PrefillCost

fn clone(&self) -> PrefillCost

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PrefillCost

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for PrefillCost

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for PrefillCost

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

where __S: Serializer,

Serialize this value into the given Serde serializer.