pub struct MtpSpeculativeParams {
pub n_max: i32,
pub n_min: i32,
pub p_min: f32,
}Expand description
Parameters for same-model MTP speculative decoding.
Fields§
§n_max: i32Maximum number of draft tokens to propose.
n_min: i32Minimum number of draft tokens required before returning a draft.
p_min: f32Minimum draft probability accepted by llama.cpp’s MTP drafter.
Trait Implementations§
Source§impl Clone for MtpSpeculativeParams
impl Clone for MtpSpeculativeParams
Source§fn clone(&self) -> MtpSpeculativeParams
fn clone(&self) -> MtpSpeculativeParams
Returns a duplicate of the value. Read more
1.0.0 (const: unstable) · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
Performs copy-assignment from
source. Read moreimpl Copy for MtpSpeculativeParams
Source§impl Debug for MtpSpeculativeParams
impl Debug for MtpSpeculativeParams
Source§impl Default for MtpSpeculativeParams
impl Default for MtpSpeculativeParams
Source§impl PartialEq for MtpSpeculativeParams
impl PartialEq for MtpSpeculativeParams
Source§fn eq(&self, other: &MtpSpeculativeParams) -> bool
fn eq(&self, other: &MtpSpeculativeParams) -> bool
Tests for
self and other values to be equal, and is used by ==.impl StructuralPartialEq for MtpSpeculativeParams
Auto Trait Implementations§
impl Freeze for MtpSpeculativeParams
impl RefUnwindSafe for MtpSpeculativeParams
impl Send for MtpSpeculativeParams
impl Sync for MtpSpeculativeParams
impl Unpin for MtpSpeculativeParams
impl UnsafeUnpin for MtpSpeculativeParams
impl UnwindSafe for MtpSpeculativeParams
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