bge-m3-fastapi

• Requirements: poetry==^1.8, python==^3.10
• Download the model.onnx and model.onnx.data files from aapot/bge-m3-onnx and move them to the models directory.

Usage

Run in local env

1. Clone this repository

   git clone https://github.com/myeolinmalchi/bge-m3-fastapi.git
   cd bge-m3-fastapi

2. Install dependencies

   poetry shell
   poetry install --no-root

3. Run main.py

   poetry run python3.10 main.py --device <device type> --batch_size <batch size>
       --device: device type for onnxruntime (cpu/cuda, default: cpu)
       --batch_size: concurrency limit for onnxruntime (default: 2)

Run with docker-compose

1. Clone this repository

   git clone https://github.com/myeolinmalchi/bge-m3-fastapi.git
   cd bge-m3-fastapi

2. Run docker-compose

   docker-compose up