Enum SleepLevel 

pub enum SleepLevel {
    L1,
    L2,
    CudaSuspend,
    Checkpoint,
    Stop,
}

Sleep level for hibernating models

Variants

L1

Level 1: Offload weights to CPU RAM (faster wake)

L2

Level 2: Discard weights (slower wake, less RAM)

CudaSuspend

Level 3: CUDA suspend via cuda-checkpoint toggle (process stays alive, GPU freed, VRAM contents held in host RAM, full state preserved)

Checkpoint

Level 4: CRIU checkpoint (snapshot process to disk, frees all GPU/CPU memory, preserves full state including KV cache, CUDA graphs, and warmed allocator)

Stop

Level 5: Stop the vLLM process entirely (full restart on wake)

Trait Implementations

impl Clone for SleepLevel

fn clone(&self) -> SleepLevel

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SleepLevel

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

Formats the value using the given formatter.

impl From<u8> for SleepLevel

fn from(level: u8) -> Self

Converts to this type from the input type.

impl PartialEq for SleepLevel

fn eq(&self, other: &SleepLevel) -> 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 Copy for SleepLevel

impl Eq for SleepLevel

impl StructuralPartialEq for SleepLevel