Enum MtpQuantization 

pub enum MtpQuantization {
    Nvfp4,
    Fp8,
    Bf16,
}

2026-09-25: Precision of the MTP head’s projection weights.

Variants

Nvfp4

Fp8

Bf16

Implementations

impl MtpQuantization

pub fn supports_drafter_prefill(self) -> bool

2026-09-25: Whether the batched drafter prefill can run at this precision. The prefill itself re-checks the weight variants and kernel handles; this lets the caller skip the max_seq_len x hidden BF16 prompt-hidden buffer for a head that could never use it. quantize_proj produces ProjectionWeight::Bf16 only for MtpQuantization::Bf16.

pub fn effective_for_head(self, dense_ffn_head: bool) -> Self

2026-09-25: The precision the head’s forward runs at (KV dtype, forward arms, drafter prefill). A dense-FFN head under Self::Nvfp4 runs the BF16 forward, so it reports Self::Bf16; every other combination reports itself. MtpHead::new, the MTP KV-pool reserve and the prompt-capture allocation all ask here.

Trait Implementations

impl Clone for MtpQuantization

fn clone(&self) -> MtpQuantization

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for MtpQuantization

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

Formats the value using the given formatter.

impl FromStr for MtpQuantization

type Err = Error

The associated error which can be returned from parsing.

fn from_str(s: &str) -> Result<Self>

Parses a string s to return a value of this type.

impl PartialEq for MtpQuantization

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

impl Eq for MtpQuantization

impl StructuralPartialEq for MtpQuantization