Press Release | Grenoble, France | April 5, 2021 helps democratize Natural Language Processing (NLP) in production launches an NLP API based on the best open-source Hugging Face and spaCy models, so any company can leverage Natural Language Processing reliably in production, and at a fair price. wants to make life easier for developers and data scientists who want to add NLP into their application.

Artificial intelligence and machine learning have made great progress these last years thanks to deep learning techniques. Natural Language Processing (NLP) is a field that experienced dramatic improvements in terms of accuracy, and many companies want to leverage NLP in their applications today.

However, even if companies invest a lot of money in NLP research projects, many projects are actually failing because of the deployment to production step. This is because creating accurate NLP models, and having them run in production, are 2 very different things. Deploying NLP models into production requires strong DevOps skills that most companies don't have.

Julien Salinas, founder and CTO at, explains: "As a software engineer, I had to face many difficulties everytime I wanted to deploy NLP models to production in a reliable and easy way. I noticed this same challenge in several projects, so I decided to start working on a robust and easy-to-integrate NLP infrastructure, in order to help developers and data scientists save weeks - even months - of tedious work." is a high performance production-ready NLP API based on the best Hugging Face and spaCy open-source models. It can be used for Named Entity Recognition (NER), sentiment-analysis, text classification, summarization, question answering, and Part-Of-Speech tagging. No DevOps is required and companies can either choose to use the ready-to-use pre-trained models, or upload the custom models developped internally by their data scientists. wants to democratize NLP by helping developers and data scientists deliver their NLP projects in no time and at a fair price.'s users leverage the API for very different use cases like for example support tickets sorting, leads extraction, social network analysis, urgency detection, public relations, economic intelligence...

A free version gives access to all the pre-trained models, up to 3 requests per minute, which is an easy way to test the quality of the models. Then higher plans grant more requests per minute, and for very demanding customers also proposes a GPU infrastructure.

In the coming months, plans to add more open-source models to its API in order to support new features and more languages. The company anticipates a 500% revenue growth and plans to hire 5 persons, before the end of the year.

About was officially launched by January 2021. By end of March it had 500 registered users. The company is based in Grenoble, France.

Press Contact

Julien Salinas, Founder and CTO at
Email: [email protected]