TextEmbed - Embedding Inference Server

TextEmbed is a high-throughput, low-latency REST API designed for serving vector embeddings. It supports a wide range of sentence-transformer models and frameworks, making it suitable for various applications in natural language processing.

Features

• High Throughput & Low Latency: Designed to handle a large number of requests efficiently.
• Flexible Model Support: Works with various sentence-transformer models.
• Scalable: Easily integrates into larger systems and scales with demand.
• Batch Processing: Supports batch processing for better and faster inference.
• OpenAI Compatible REST API Endpoint: Provides an OpenAI compatible REST API endpoint.
• Single Line Command Deployment: Deploy multiple models via a single command for efficient deployment.
• Support for Embedding Formats: Supports binary, float16, and float32 embeddings formats for faster retrieval.

Getting Started

Prerequisites

Ensure you have Python 3.10 or higher installed. You will also need to install the required dependencies.

Installation

1. Install the required dependencies:

   pip install -U textembed

2. Start the TextEmbed server with your desired models:

   python3 -m textembed.server --models <Model1>, <Model2> --port <Port>

   Replace <Model1> and <Model2> with the names of the models you want to use, separated by commas. Replace <Port> with the port number on which you want to run the server.

For more information about the Docker deployment and configuration, please refer to the documentation setup.md.