Enum LlamaFlashAttnType 

#[repr(i32)]
pub enum LlamaFlashAttnType {
    Auto = -1,
    Disabled = 0,
    Enabled = 1,
}

Flash-attention enablement policy for the context.

Maps to llama_context_params.flash_attn_type. LlamaFlashAttnType::Auto lets llama.cpp pick based on model architecture and active backend.

Variants

Auto = -1

Let llama.cpp decide based on model and backend.

Disabled = 0

Force flash attention off.

Enabled = 1

Force flash attention on.

Trait Implementations

impl Clone for LlamaFlashAttnType

fn clone(&self) -> LlamaFlashAttnType

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for LlamaFlashAttnType

impl Debug for LlamaFlashAttnType

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

Formats the value using the given formatter.

impl Eq for LlamaFlashAttnType

impl From<LlamaFlashAttnType> for i32

fn from(value: LlamaFlashAttnType) -> Self

Converts to this type from the input type.

impl From<i32> for LlamaFlashAttnType

fn from(value: i32) -> Self

Converts to this type from the input type.

impl PartialEq for LlamaFlashAttnType

fn eq(&self, other: &LlamaFlashAttnType) -> 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 StructuralPartialEq for LlamaFlashAttnType