Struct PredictiveOffloadInfo 

pub struct PredictiveOffloadInfo {
    pub enabled: bool,
    pub gpu_expert_budget_per_layer: usize,
    pub num_sparse_moe_layers: usize,
    pub expert_param_bytes: usize,
    pub cuda_free_bytes: Option<usize>,
    pub cuda_total_bytes: Option<usize>,
    pub reserve_bytes: usize,
    pub jump_steps: usize,
    pub collect_stats: bool,
}

Return value of TIDE enable_predictive_expert_offload (JSON-serializable keys).

Fields

enabled: bool

gpu_expert_budget_per_layer: usize

num_sparse_moe_layers: usize

expert_param_bytes: usize

cuda_free_bytes: Option<usize>

cuda_total_bytes: Option<usize>

reserve_bytes: usize

jump_steps: usize

collect_stats: bool

Trait Implementations

impl Clone for PredictiveOffloadInfo

fn clone(&self) -> PredictiveOffloadInfo

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for PredictiveOffloadInfo

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

Formats the value using the given formatter.

impl PartialEq for PredictiveOffloadInfo

fn eq(&self, other: &PredictiveOffloadInfo) -> 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 Eq for PredictiveOffloadInfo

impl StructuralPartialEq for PredictiveOffloadInfo