pub struct CacheMeta {
pub page_size_tokens: u32,
pub n_layers: u32,
pub n_heads: u32,
pub head_dim: u32,
pub dtype: Dtype,
}Expand description
Static metadata describing a paged KV cache. Identical across pages of
the same engine instance; embedded once in the PageManifest.
Fields§
§page_size_tokens: u32Tokens per page (vLLM default 16).
n_layers: u32Number of transformer layers.
n_heads: u32Number of attention heads.
head_dim: u32Per-head dimension.
dtype: DtypeNumeric dtype.
Implementations§
Trait Implementations§
Source§impl<'de> Deserialize<'de> for CacheMeta
impl<'de> Deserialize<'de> for CacheMeta
Source§fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
Deserialize this value from the given Serde deserializer. Read more
impl Copy for CacheMeta
impl Eq for CacheMeta
impl StructuralPartialEq for CacheMeta
Auto Trait Implementations§
impl Freeze for CacheMeta
impl RefUnwindSafe for CacheMeta
impl Send for CacheMeta
impl Sync for CacheMeta
impl Unpin for CacheMeta
impl UnsafeUnpin for CacheMeta
impl UnwindSafe for CacheMeta
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