pub struct Fp8DenseWeight {
pub weight: DevicePtr,
pub row_scale: DevicePtr,
}Expand description
2026-09-25: FP8 E4M3 weight quantized from BF16 at load, with one f32 scale per row.
Fields§
§weight: DevicePtr2026-09-25: [N, K] FP8 E4M3 bytes.
row_scale: DevicePtr2026-09-25: [N] f32 row scales.
Trait Implementations§
Source§impl Clone for Fp8DenseWeight
impl Clone for Fp8DenseWeight
Source§fn clone(&self) -> Fp8DenseWeight
fn clone(&self) -> Fp8DenseWeight
Returns a duplicate of the value. Read more
1.0.0 · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
Performs copy-assignment from
source. Read moreSource§impl Debug for Fp8DenseWeight
impl Debug for Fp8DenseWeight
impl Copy for Fp8DenseWeight
Auto Trait Implementations§
impl Freeze for Fp8DenseWeight
impl RefUnwindSafe for Fp8DenseWeight
impl Send for Fp8DenseWeight
impl Sync for Fp8DenseWeight
impl Unpin for Fp8DenseWeight
impl UnwindSafe for Fp8DenseWeight
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more