Vladimir Araujo

Vladimir Araujo

AI and NLP Researcher

Posts by Tags

LatamGPT

Local Knowledge at a Price: Reading LatamGPT Through Its Benchmarks

10 minute read

Published:

In February 2026, LatamGPT was presented to a wave of regional enthusiasm: the first open large language model built from and for Latin America and the Caribbean, coordinated by Chile's CENIA and more than sixty institutions across the region. The promise was sovereignty and cultural relevance—moving Latin America from a consumer of foreign models to a producer of its own. When the model's open weights were released on June 1, 2026, though, they arrived without a single benchmark number. As someone who has spent years working with language models, that absence nagged at me. So I decided to run the comparison myself, and to ask the question the release skipped: what did all that training actually buy, and what did it cost?

Spanish NLP

Local Knowledge at a Price: Reading LatamGPT Through Its Benchmarks

10 minute read

Published:

In February 2026, LatamGPT was presented to a wave of regional enthusiasm: the first open large language model built from and for Latin America and the Caribbean, coordinated by Chile's CENIA and more than sixty institutions across the region. The promise was sovereignty and cultural relevance—moving Latin America from a consumer of foreign models to a producer of its own. When the model's open weights were released on June 1, 2026, though, they arrived without a single benchmark number. As someone who has spent years working with language models, that absence nagged at me. So I decided to run the comparison myself, and to ask the question the release skipped: what did all that training actually buy, and what did it cost?

Democratizing Spanish NLP: Language Models and Benchmarks

5 minute read

Published:

In recent years, the field of Natural Language Processing (NLP) has witnessed significant advancements, particularly in the development of pre-trained language models. While English has been at the forefront of these developments, there has been a growing need to extend these advancements to other languages, including Spanish. Addressing this gap, our research focuses on creating and evaluating Spanish language models and resources that are not only effective but also accessible to a broader community. This blog post delves into our efforts in developing lightweight models, establishing evaluation benchmarks, and introducing sequence-to-sequence models tailored for the Spanish language.

benchmarks

Local Knowledge at a Price: Reading LatamGPT Through Its Benchmarks

10 minute read

Published:

In February 2026, LatamGPT was presented to a wave of regional enthusiasm: the first open large language model built from and for Latin America and the Caribbean, coordinated by Chile's CENIA and more than sixty institutions across the region. The promise was sovereignty and cultural relevance—moving Latin America from a consumer of foreign models to a producer of its own. When the model's open weights were released on June 1, 2026, though, they arrived without a single benchmark number. As someone who has spent years working with language models, that absence nagged at me. So I decided to run the comparison myself, and to ask the question the release skipped: what did all that training actually buy, and what did it cost?

Democratizing Spanish NLP: Language Models and Benchmarks

5 minute read

Published:

In recent years, the field of Natural Language Processing (NLP) has witnessed significant advancements, particularly in the development of pre-trained language models. While English has been at the forefront of these developments, there has been a growing need to extend these advancements to other languages, including Spanish. Addressing this gap, our research focuses on creating and evaluating Spanish language models and resources that are not only effective but also accessible to a broader community. This blog post delves into our efforts in developing lightweight models, establishing evaluation benchmarks, and introducing sequence-to-sequence models tailored for the Spanish language.

continued pre-training

Local Knowledge at a Price: Reading LatamGPT Through Its Benchmarks

10 minute read

Published:

In February 2026, LatamGPT was presented to a wave of regional enthusiasm: the first open large language model built from and for Latin America and the Caribbean, coordinated by Chile's CENIA and more than sixty institutions across the region. The promise was sovereignty and cultural relevance—moving Latin America from a consumer of foreign models to a producer of its own. When the model's open weights were released on June 1, 2026, though, they arrived without a single benchmark number. As someone who has spent years working with language models, that absence nagged at me. So I decided to run the comparison myself, and to ask the question the release skipped: what did all that training actually buy, and what did it cost?

data mining

Recommendation Systems for Item Recommendation in MOBA Games

10 minute read

Published:

The video game industry has adopted recommendation systems to boost users' interest with a focus on game sales. Other exciting applications within video games are those that help the player to make decisions that would maximize their gaming experience. In this blog, I am going to present to you a research focused on the second application that resulted in two papers presented in RecSys.

deep learning

Recommendation Systems for Item Recommendation in MOBA Games

10 minute read

Published:

The video game industry has adopted recommendation systems to boost users' interest with a focus on game sales. Other exciting applications within video games are those that help the player to make decisions that would maximize their gaming experience. In this blog, I am going to present to you a research focused on the second application that resulted in two papers presented in RecSys.

item recommendation

Recommendation Systems for Item Recommendation in MOBA Games

10 minute read

Published:

The video game industry has adopted recommendation systems to boost users' interest with a focus on game sales. Other exciting applications within video games are those that help the player to make decisions that would maximize their gaming experience. In this blog, I am going to present to you a research focused on the second application that resulted in two papers presented in RecSys.

language models

Local Knowledge at a Price: Reading LatamGPT Through Its Benchmarks

10 minute read

Published:

In February 2026, LatamGPT was presented to a wave of regional enthusiasm: the first open large language model built from and for Latin America and the Caribbean, coordinated by Chile's CENIA and more than sixty institutions across the region. The promise was sovereignty and cultural relevance—moving Latin America from a consumer of foreign models to a producer of its own. When the model's open weights were released on June 1, 2026, though, they arrived without a single benchmark number. As someone who has spent years working with language models, that absence nagged at me. So I decided to run the comparison myself, and to ask the question the release skipped: what did all that training actually buy, and what did it cost?

Democratizing Spanish NLP: Language Models and Benchmarks

5 minute read

Published:

In recent years, the field of Natural Language Processing (NLP) has witnessed significant advancements, particularly in the development of pre-trained language models. While English has been at the forefront of these developments, there has been a growing need to extend these advancements to other languages, including Spanish. Addressing this gap, our research focuses on creating and evaluating Spanish language models and resources that are not only effective but also accessible to a broader community. This blog post delves into our efforts in developing lightweight models, establishing evaluation benchmarks, and introducing sequence-to-sequence models tailored for the Spanish language.

recommendation system

Recommendation Systems for Item Recommendation in MOBA Games

10 minute read

Published:

The video game industry has adopted recommendation systems to boost users' interest with a focus on game sales. Other exciting applications within video games are those that help the player to make decisions that would maximize their gaming experience. In this blog, I am going to present to you a research focused on the second application that resulted in two papers presented in RecSys.

seq2seq

Democratizing Spanish NLP: Language Models and Benchmarks

5 minute read

Published:

In recent years, the field of Natural Language Processing (NLP) has witnessed significant advancements, particularly in the development of pre-trained language models. While English has been at the forefront of these developments, there has been a growing need to extend these advancements to other languages, including Spanish. Addressing this gap, our research focuses on creating and evaluating Spanish language models and resources that are not only effective but also accessible to a broader community. This blog post delves into our efforts in developing lightweight models, establishing evaluation benchmarks, and introducing sequence-to-sequence models tailored for the Spanish language.