Struct LocalFastembedProvider 

pub struct LocalFastembedProvider { /* private fields */ }

EmbeddingProvider backed by a fastembed-managed ONNX model.

Implementations

impl LocalFastembedProvider

pub fn new(key: &str, cache_dir: impl AsRef<Path>) -> Result<Self>

Build a provider for the curated key (see registry::REGISTRY). Downloads the model on first use; subsequent runs read from cache.

pub fn cache_dir(&self) -> &Path

Where the model files are cached.

pub fn model_info(&self) -> &'static CuratedModel

The curated model entry this provider serves.

Trait Implementations

impl Debug for LocalFastembedProvider

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

Formats the value using the given formatter.

impl EmbeddingProvider for LocalFastembedProvider

fn model_id(&self) -> ModelId

Stable id — see invariants in the module docs.

fn dim(&self) -> u16

Vector dimensionality. Must match every vector returned.

fn embed_batch<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, texts: &'life1 [&'life2 str], task: EmbeddingTask, ) -> Pin<Box<dyn Future<Output = Result<Vec<Vec<f32>>>> + Send + 'async_trait>>

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

Embed a batch of texts. The provider is responsible for chunking the batch into model-friendly sizes; the worker calling this owns the outer batching loop.

fn embed_query<'life0, 'life1, 'async_trait>( &'life0 self, text: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<Vec<f32>, Error>> + Send + 'async_trait>>

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

Embed a single query string. Default impl forwards to embed_batch.