Header abstract shape backgroud 1Header abstract shape backgroud 2Header abstract shape backgroud 3Header abstract shape backgroud 4Header abstract shape backgroud 5

Babelscape

Language generation, semantically grounded
Customers & Partners:
  • Adobe
  • ATEX
  • EUIPO
  • Oxford University Press
  • University of Florida
Header shape illustration 1Header shape illustration 2Header shape illustration 2

Minerva: the first Italian LLM

Minerva is the first LLM created from scratch with a focus on Italian, trained on an extensive dataset of open-access Italian and English sources, comprising trillions of words. Developed by the Sapienza NLP group, led by Professor Roberto Navigli, Minerva represents a groundbreaking advance in multilingual natural language processing. In collaboration with CINECA’s Leonardo supercomputer, Minerva benefits from state-of-the-art hardware.

Babelscape is spearheading the industrialization of Minerva, optimizing the model for real-world applications that require high-performance AI solutions. Minerva will soon be available to businesses, offering transparent and robust capabilities in text understanding, generation, and translation.

Minerva LLM logo

Find out how our solutions fit your needs

Where can we apply our technologies for you?
Each business is unique, which is why we customize our products to support your growth.

If you don’t see your industry listed, contact us to explore tailored solutions for you!

Contact us

Publishing and media

Babelscape enhances media content management by processing multilingual newspaper articles and providing semantic tagging. This supports semantic search for users and assists journalists in annotating and linking their work across languages.

Pharmaceutical

We assist pharmaceutical companies in harnessing knowledge from diverse sources, facilitating competitor monitoring and insight gathering for product development and strategic initiatives.

Market Research

Discover consumer perceptions and market trends through our advanced analysis tools, which deliver precise insights across multiple languages, empowering your market research strategies.

Social Media

Social media platforms are rich with user opinions. Babelscape simplifies the extraction and summarization of multilingual content from these platforms, turning vast amounts of data into manageable and actionable insights.

Law, patents and trademarks

We provide specialized solutions for helping officials and consultants in the pre-registration process, by checking the validity of a trademark or patent. We also offer solutions for building law-specific semantic search engines.

Why Babelscape?

Multilinguality and Semantic Grounding

At Babelscape, we excel in delivering multilingual solutions that transcend the capabilities of traditional models by combining advanced large language models with our state-of-the-art multilingual knowledge graph, WordAtlas.

Our systems are designed to capture, interpret and interconnect a wide range of concepts, entities, relations and emotions.

Multilinguality and Semantic Grounding illustration

Performance beyond the state of the art

Babelscape sets new benchmarks in NLP performance, consistently delivering results that surpass the current state of the art.

Our innovative algorithms and continuous research efforts enable us to tackle complex language processing challenges more effectively, providing our clients with superior tools that stay ahead of technological advancements.

Performance beyond the state of the art illustration

Secure data management

We prioritize the security and integrity of your data. Babelscape employs advanced encryption and rigorous data protection protocols to safeguard your information at every step of the processing chain.

This secure approach ensures that your data remains confidential, compliant, and shielded from vulnerabilities, providing you with a reliable foundation for data-driven decision-making.

Secure data management illustration
Find out more
Why Babelscape layerMultilinguality and Semantic Grounding backgroundPerformance beyond the state of the art backgroundSecure data management background

Babelscape is a proud sponsor of the top-tier conference in Natural Language Processing ACL 2025.

Research & Publications

Babelscape is committed to carrying out research and develop technological innovation at the highest level in the field of multilingual Natural Language Understanding with a strong emphasis on neuro-symbolic approaches which combine multilingual knowledge graphs with deep learning models.

Out latest publications:

LLM
Francesco Maria Molfese, Luca Moroni, Luca Gioffrè, Alessandro Scirè, Simone Conia, Roberto Navigli
Right Answer, Wrong Score: Uncovering the Inconsistencies of LLM Evaluation in Multiple-Choice Question Answering
Findings of ACL 2025
T. Nakamura, M. Mishra, S. Tedeschi, Y. Chai, J. Stillerman, F. Friedrich, P. Yadav, T. Laud, V. Chien, T. Zhuo and 35 more
Aurora-M: The First Open Source Multilingual Language Model Red-teamed according to the U.S. Executive Order
Proceedings of COLING 2025

What’s new at Babelscape

See all
Card preview illustration
Press Release

Graph RAG: smarter AI retrieval through knowledge graphs

Retrieval-Augmented Generation (RAG) has revolutionized how AI systems access information, but it still struggles with context limitations and factual hallucinations.

At Babelscape, we have been experimenting with this technology on our multilingual knowledge graph WordAtlas (and others!) for quite some time now, with the aim of demonstrating how Graph RAG can transform AI understanding across languages.

Card preview illustration
Product Upgrade

Babelscape Vera is now live - Try the free demo!

We’re thrilled to announce that Babelscape Vera, our LLM-powered virtual assistant for fact-checking, is now online! Vera intelligently gathers, analyzes, and validates claims using credible sources, ensuring accuracy and transparency through in-depth, evidence-based analysis.
The live demo is now available, try it for free and experience the future of fact-checking today!

Card preview illustration
Press Release

Minerva: Italy's First Family of Large Language Models trained on Italian texts

The Minerva family of large language models (LLMs) is reshaping the AI landscape in Italy. Created by Sapienza NLP in collaboration with FAIR (Future Artificial Intelligence Research) and CINECA, with additional support from Babelscape, Minerva is the first suite of AI models built from scratch to serve Italian-language needs while also supporting English.

Your privacy choices

Save and continue
Sign up!
The best way to get the latest news from Babelscape and the NLP world!
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Thank you for subscribing!
You’ve been added to our mailing list, and you’ll receive our next newsletter to stay updated on the latest news from the NLP world!
Something went wrong
We are sorry, your request cannot be processed right now.
Please wait a bit and try again.
Unsubscribe
We're sorry to see you go. Please enter your email address to complete the unsubscription process.
You'll receive an email confirmation shortly.
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Check your email
We have sent you a link to your email to complete the unsubscribe process.
Something went wrong
We are sorry, your request cannot be processed right now.
Please wait a bit and try again.