Struct KvCacheEstimateRequest 

pub struct KvCacheEstimateRequest {
    pub engine: EngineConfig,
    pub max_num_tokens: u32,
    pub max_batch_size: u32,
    pub kv_cache_memory_fraction: KvCacheMemoryFraction,
    pub gpu_memory_capacity_bytes_override: Option<u64>,
    pub tolerance_fraction: Option<f64>,
    pub options: KvCacheEstimateOptions,
}

KV-cache memory request. engine reuses the modularised EngineConfig; the remaining fields describe the runtime sizing budget.

Fields

engine: EngineConfig

max_num_tokens: u32

max_batch_size: u32

kv_cache_memory_fraction: KvCacheMemoryFraction

gpu_memory_capacity_bytes_override: Option<u64>

Override for unknown SKUs; when Some, it wins over the SystemSpec capacity reported by the native path.

tolerance_fraction: Option<f64>

None = raw estimate only; Some(0.05) = 5% safety margin.

options: KvCacheEstimateOptions

Trait Implementations

impl Clone for KvCacheEstimateRequest

fn clone(&self) -> KvCacheEstimateRequest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for KvCacheEstimateRequest

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for KvCacheEstimateRequest

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for KvCacheEstimateRequest

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for KvCacheEstimateRequest

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 KvCacheEstimateRequest