Struct Model2VecProvider 

pub struct Model2VecProvider { /* private fields */ }

Model2Vec embedding provider for fast embeddings.

Loads the model into memory on creation for instant inference. Typical latency: < 1ms per embedding.

Implementations

impl Model2VecProvider

pub fn new() -> Result<Self>

Create a new Model2Vec provider with the default model (potion-base-8M).

Errors

Returns an error if the model cannot be loaded from HuggingFace Hub.

pub fn with_model(model_name: Option<String>) -> Result<Self>

Create a new Model2Vec provider with a custom model.

Arguments

• model_name - Optional model name. Defaults to minishlab/potion-base-8M.

Errors

Returns an error if the model cannot be loaded.

pub fn try_new() -> Option<Self>

Try to create a provider, returning None if model loading fails.

Useful for graceful fallback when Model2Vec isn’t available.

Trait Implementations

impl EmbeddingProvider for Model2VecProvider

fn info(&self) -> ProviderInfo

Get provider metadata.

async fn is_available(&self) -> bool

Check if the provider is available.

async fn generate_embedding(&self, text: &str) -> Result<Vec<f32>>

Generate embedding for a single text.

async fn generate_embeddings(&self, texts: &[&str]) -> Result<Vec<Vec<f32>>>

Generate embeddings for multiple texts (batch).