---
title: README
emoji: 📊
colorFrom: red
colorTo: yellow
sdk: static
pinned: false
---

## Hello, we're Minish!

We're a two-person ([@pringled](https://github.com/Pringled) and [@stephantul](https://github.com/stephantul)) open-source lab, with a focus on Natural Language Processing.

We believe that if you make models fast enough, you unlock new possibilities.

Using our software, you can:
* Embed the entire English Wikipedia in 5 minutes
* Classify tens of thousands of documents per second on a CPU
* Approximately deduplicate extremely large datasets in minutes
* Build the fastest RAG application in the world
* Easily evaluate which ANN algorithm works best for your data

Our projects:

* [model2vec](https://github.com/MinishLab/model2vec): tiny static embedding models with state-of-the-art performance.
* [potion](https://huggingface.co/collections/minishlab/potion-6721e0abd4ea41881417f062): the best small models in the world. 100-500x faster than a sentence-transformer, and almost as good.
* [vicinity](https://github.com/MinishLab/vicinity): consistent interfaces to many approximate nearest neighbor algorithms.
* [semhash](https://github.com/MinishLab/semhash): lightning-fast, super accuracte, semantic deduplication and filtering for your text datasets.
* [model2vec-rs](https://github.com/MinishLab/model2vec-rs): a Rust port of model2vec.



You can also find us on: 
🔬 [GitHub](https://github.com/MinishLab)
👽 [LinkedIn](https://www.linkedin.com/company/minish-lab/)
💬 [Discord](https://discord.gg/4BDPR5nmtK)