POTION-BASE-8M

Introduction

POTION-BASE-8M is a model hosted on Hugging Face by MINISHLAB. It offers various capabilities including embeddings and static-embeddings, suitable for different types of data processing and model evaluation tasks.

Architecture

The model is built with support for ONNX and Safetensors, providing flexibility in deployment and security for model weights. It is designed to work with Model2Vec and is compatible with mteb for benchmarking.

Training

Details regarding the training process are not provided in the given document. For specific training details, users are encouraged to explore the model's repository or associated documentation.

Guide: Running Locally

To run POTION-BASE-8M locally, follow these steps:

1. Clone the repository from Hugging Face:

   git clone https://huggingface.co/minishlab/potion-base-8M
   

2. Navigate to the cloned directory:

   cd potion-base-8M
   

3. Install the required dependencies:

   pip install -r requirements.txt
   

4. Run the model using your preferred environment.

For optimal performance, especially with large datasets, consider using cloud GPUs such as those offered by AWS, Google Cloud, or Azure.

License

This model is released under the MIT License, allowing for broad use, modification, and distribution.