VectraYX-Nano: A 42M-Parameter Spanish Cybersecurity Language Model with Curriculum Learning and Native Tool Use Conference: Preprint; 2026; CCS: Computing methodologies Natural language generation CCS: Security and privacy Software and application security CCS: Computing methodologies Neural networks
Juan S. Santillana · 2026
Read VectraYX-Nano: A 42M-Parameter Spanish Cybersecurity Language Model with Curriculum Learning and Native Tool Use Conference: Preprint; 2026; CCS: Computing methodologies Natural language generation CCS: Security and privacy Software and application security CCS: Computing methodologies Neural networks by Juan S. Santillana (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-13989.txt.
Introduction
Santillana offers VectraYX-Nano, a decoder-only language model of about 42 million weights, trained from scratch in Spanish for cyber defense. It keeps a Latin-American focus and calls tools natively via the Model Context Protocol. Its text set, VectraYX-Sec-ES, holds 170 million tokens, built by an eight-machine line for about 25 dollars of cloud compute. The set splits into chat, cyber defense and attack-tool phases.
Catalog record
| CREATOR | Juan S. Santillana |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 128153 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.13989