Enum KernelPrecision 

pub enum KernelPrecision {
    FP32,
    FP16,
    BF16,
    FP8,
}

Precision mode for kernel computations.

Variants

FP32

FP32 throughout (highest precision, slowest).

FP16

FP16 compute with FP32 accumulation (balanced).

BF16

BF16 compute with FP32 accumulation (best for training).

FP8

FP8 compute with FP16 accumulation (fastest, Hopper+).

Trait Implementations

impl Clone for KernelPrecision

fn clone(&self) -> KernelPrecision

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for KernelPrecision

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

Formats the value using the given formatter.

impl Default for KernelPrecision

fn default() -> KernelPrecision

Returns the “default value” for a type.

impl PartialEq for KernelPrecision

fn eq(&self, other: &KernelPrecision) -> 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 KernelPrecision

impl Eq for KernelPrecision

impl StructuralPartialEq for KernelPrecision