Crate lumosai_vector_fastembed 

LumosAI FastEmbed Integration

This crate provides FastEmbed integration for LumosAI vector storage, enabling local embedding generation without external API dependencies.

Features

• Local Processing: Generate embeddings locally without API calls
• Multiple Models: Support for various pre-trained models
• High Performance: Optimized for batch processing
• Easy Integration: Seamless integration with LumosAI vector storage

Quick Start

use lumosai_vector_fastembed::{FastEmbedProvider, FastEmbedModel};
use lumosai_vector_core::traits::EmbeddingModel;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create FastEmbed provider
    let provider = FastEmbedProvider::new(FastEmbedModel::BGESmallENV15).await?;
     
    // Generate embedding
    let embedding = provider.embed_text("Hello, world!").await?;
    println!("Embedding dimensions: {}", embedding.len());
     
    Ok(())
}

Re-exports

pub use models::FastEmbedModel;

pub use models::ModelInfo;

pub use provider::FastEmbedProvider;

pub use error::FastEmbedError;

pub use error::Result;

Modules

error

Error types for FastEmbed integration

models

FastEmbed model definitions and metadata

provider

FastEmbed embedding provider implementation

Structs

FastEmbedClient

FastEmbed client for managing embedding models

FastEmbedConfig

Configuration for FastEmbed client

FastEmbedConfigBuilder

Builder for FastEmbed configuration

Traits

EmbeddingModel

Trait for embedding models

Type Aliases

Metadata

Metadata type for storing arbitrary key-value pairs

Vector

Vector type alias for f32 vectors