CEO's Column
Search
More
Foundation Models

Latam-GPT: A Collaborative AI Language Model for Latin America

ByNeelima N M
2025-06-18.14 days ago
Latam-GPT: A Collaborative AI Language Model for Latin America
Twelve Latin American countries join forces to develop Latam-GPT, an open-source AI model tailored to the region’s languages, cultures, and Indigenous heritage.

Twelve Latin American countries have announced their collaboration on Latam-GPT, a groundbreaking artificial intelligence language model set to launch in September, as reported by Reuters.

The model, designed to understand and cater to the linguistic and cultural diversity of the region, is being spearheaded by Chile's state-run National Center for Artificial Intelligence (CENIA), in partnership with over 30 regional institutions.

Democratizing AI and Reflecting Regional Diversity

Latam-GPT is being developed as an open-source project to address the limitations of global AI models, which are primarily trained on English and often fail to capture the nuances of Latin American languages and cultures.

Chilean Science Minister Aisen Etcheverry described the project as a potential "democratizing element for AI," aiming to increase the reach and use of AI across Latin America, particularly in educational and healthcare settings.

The AI model is designed to better reflect the local cultures, offering more accurate and context-aware responses compared to existing models.

Fostering Cultural Preservation and Linguistic Inclusivity

According to Reuters, a core objective of Latam-GPT is to preserve and promote Indigenous languages. As part of this effort, an initial translation tool for Rapa Nui, the language of Easter Island, has already been developed.

The project plans to expand this to other Indigenous languages, enabling applications like virtual public service assistants and personalized education systems that respect and include regional languages.

Also read: Patria Launches Omnia to Build Green AI Data Centers in Latin America

Regional Collaboration and Technological Foundation

The Latam-GPT model is built on Llama 3 AI technology and is being trained through a network of regional computers, including facilities at Chile's University of Tarapaca and cloud-based systems supported by Amazon Web Services. The initiative is also backed by the regional development bank CAF, further solidifying its regional importance.

Reuters reports that while the project is still in its early stages and lacks a dedicated budget, CENIA head Alvaro Soto is optimistic that the model's capabilities will attract further funding and support. Soto highlighted that Latam-GPT is not intended to compete with consumer products like ChatGPT but instead aims to create a practical tool for developing AI applications tailored to the region's unique needs.

Related Topics

Large Language Models (LLMs)Foundation Models

Subscribe to NG.ai News for real-time AI insights, personalized updates, and expert analysis—delivered straight to your inbox.