Struct AutomaticPlannerPolicy 

#[non_exhaustive]
pub struct AutomaticPlannerPolicy {
    pub device_memory_fallback_bytes: u64,
    pub host_memory_fallback_bytes: u64,
    pub memory_headroom_percent: u8,
    pub expert_cache_share_percent: u8,
    pub device_layer_window: usize,
    pub max_cached_shards: usize,
    pub embedded_mtp_draft_tokens: usize,
    pub minimum_feedback_tokens: usize,
}

Tunable, serializable automatic-planning policy.

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

device_memory_fallback_bytes: u64

Device budget used when current device availability is unavailable.

host_memory_fallback_bytes: u64

Host budget used when current host availability is unavailable.

memory_headroom_percent: u8

Percentage of observed free memory reserved for runtime state and drift.

expert_cache_share_percent: u8

Percentage of bounded residency budgets assigned to routed experts.

device_layer_window: usize

Repeated execution groups retained in the layerwise device window.

max_cached_shards: usize

Maximum simultaneously cached checkpoint shards or readers.

embedded_mtp_draft_tokens: usize

Maximum proposals used when embedded MTP is available.

minimum_feedback_tokens: usize

Minimum generated-token count for one prior run to influence planning.

Trait Implementations

impl Clone for AutomaticPlannerPolicy

fn clone(&self) -> AutomaticPlannerPolicy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AutomaticPlannerPolicy

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

Formats the value using the given formatter.

impl Default for AutomaticPlannerPolicy

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for AutomaticPlannerPolicy

where AutomaticPlannerPolicy: Default,

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for AutomaticPlannerPolicy

impl PartialEq for AutomaticPlannerPolicy

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

Equality operator ==.

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

Inequality operator !=.

impl Serialize for AutomaticPlannerPolicy

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 AutomaticPlannerPolicy