/ Blog /
News
ALIA's Tokenizer: Why Spanish, Catalan, Basque, and Galician Cost Almost Half the Tokens of Llama 3
ALIA tokenizes Spanish, Catalan, Basque, and Galician with ~1.7× fewer tokens than Llama 3 or Mistral. We demonstrate this with an interactive demo running 100% in your browser. Here's why it matters: context, cost, and quality.
Read more →
ALIA on NVIDIA DGX Spark: Adapt Spain's State LLM to Your Domain on a mini HPC
We quantized ALIA, the Spanish Government's AI model, to NVFP4—NVIDIA Blackwell's native format. What matters: it now fits comfortably in an NVIDIA DGX Spark with room for local fine-tuning—adapting ALIA to your terminology, your documents, your domain. Now available on Hugging Face.
Read more →
OpenAI Privacy Filter: Protect Your Data Without Sending It to the Cloud
Discover how Privacy Filter, OpenAI's model for detecting and redacting personal data, works. Montevive presents a 100% local browser demo: real privacy without sending information to servers.
Read more →
Stanford AI Index 2026: AI Accelerates, But Who's in Control?
The Stanford AI Index 2026 confirms: AI advances at record speed but safety, transparency, and regulation lag behind. Incidents increased 55% in one year. Does your company control its AI, or is it controlled by it?
Read more →
AI in the Startup's DNA: Chema Robles at the El Carmen Incubator
Our CEO Chema Robles gave a 90-minute talk on artificial intelligence in startups during the welcome session of the second cohort of the El Carmen Incubator.
Read more →
Live Demo: Prompt Injection in AI Agents with n8n
We have created a practical demonstration of prompt injection vulnerabilities in local AI agents. Discover how a model can be manipulated to exfiltrate sensitive information and which models better resist these attacks.
Read more →
AI and Employment: What Anthropic Says (and What They're Not Telling You)
Anthropic has published the most detailed map to date of which jobs AI is replacing. And the most interesting data point isn't in the headlines. It's not that programmers are the most exposed (they are, with a...
Read more →
Secure AI in 2026: 5 Trends Every Entrepreneur Should Know and How to Prepare
2025 was the year of AI experimentation. 2026 will be the year of consequences. I'll tell you what's coming and how to prepare.
Read more →
2026: The Year AI Must Demonstrate Real Value, Security, and Efficiency
While the industry demands ROI within 6 months, traditional security frameworks are failing, and Europe is waking up late to its technological dependency. Welcome to the year where AI must move from promises to facts.
Read more →[ Let's talk ]
The risk you don't see, and the opportunity you're not seizing yet
We reply in less than 24 hours.
Request your diagnostic
We'll reply within 24h.