EmbedCache¶

High-performance text embedding library and service with intelligent caching

EmbedCache is a Rust library and REST API service that provides fast text embedding generation with built-in caching capabilities. It supports multiple state-of-the-art embedding models and offers flexible text chunking strategies, including LLM-powered intelligent chunking.

Key Features¶

Multiple Embedding Models - Support for 22+ models including BGE, MiniLM, Nomic, and Multilingual E5
Intelligent Chunking - Word-based, LLM concept-based, and LLM introspection-based chunking strategies
Built-in Caching - SQLite-based caching to avoid redundant embedding computations
Dual Interface - Use as a Rust library or deploy as a REST API service
Extensible - Easy to add custom chunking strategies and embedding providers
High Performance - Async/await architecture with efficient batch processing

Quick Start¶

As a Service¶

# Install
cargo install embedcache

# Run
embedcache

As a Library¶

use embedcache::{FastEmbedder, Embedder};
use fastembed::{InitOptions, EmbeddingModel};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let embedder = FastEmbedder {
        options: InitOptions::new(EmbeddingModel::BGESmallENV15),
    };

    let texts = vec![
        "Hello, world!".to_string(),
        "EmbedCache is awesome!".to_string(),
    ];

    let embeddings = embedder.embed(&texts).await?;
    println!("Generated {} embeddings", embeddings.len());
    Ok(())
}

Use Cases¶

Semantic Search - Generate embeddings for search indexing and retrieval
RAG Applications - Process documents for retrieval-augmented generation
Text Similarity - Compare text documents using embedding similarity
Content Deduplication - Identify similar or duplicate content
Caching Layer - Reduce embedding API costs with intelligent caching

Getting Started¶

Ready to get started? Check out the Installation Guide or jump straight to the Quick Start Tutorial.

License¶

EmbedCache is released under the GPL-3.0 license. See the LICENSE file for details.