Struct SqliteVecBackend 

pub struct SqliteVecBackend { /* private fields */ }

Exact-KNN vector backend over a vec0 virtual table.

Implementations

impl SqliteVecBackend

pub fn open(path: &str, dim: usize) -> Result<Self>

Open (or create) a backend at path (":memory:" for in-RAM) with the given fixed dimension. The vec0 virtual table is created if absent.

pub fn in_memory(dim: usize) -> Result<Self>

In-RAM backend (shorthand for open(":memory:", dim)).

Trait Implementations

impl VectorBackend for SqliteVecBackend

fn upsert<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, id: &'life1 str, embedding: &'life2 [f32], metadata: Metadata, ) -> Pin<Box<dyn Future<Output = Result<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Upsert a vector with metadata. If id exists it is replaced.

fn search<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, query: &'life1 [f32], top_k: usize, filter: Option<&'life2 Filter>, ) -> Pin<Box<dyn Future<Output = Result<Vec<VectorHit>>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait,

Search for vectors similar to query, returning at most top_k hits that satisfy filter (pre-filter where supported).

fn delete<'life0, 'life1, 'async_trait>( &'life0 self, id: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<()>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait, 'life1: 'async_trait,

Delete a vector by ID. No-op if absent.

fn dimension(&self) -> usize

The fixed dimension of vectors this backend accepts.