Blog
Notes and analysis on AI
Trends, tutorials, and opinions from the Lixto Labs team on AI applied to business.
GPT-5 and extended reasoning models: what changes for businesses in 2026
A practical look at how GPT-5 and its extended reasoning modes are redefining what companies can automate with AI.
GPT-5Reasoning modelsStrategyAutonomous AI agents in production: lessons from the past year
After 12 months running AI agents with real clients, here's what works, what fails, and how to avoid the most common pitfalls.
AI AgentsProductionLessons learnedClaude 4 vs GPT-5 vs Gemini 2.5: a B2B comparison
Which model fits which enterprise task in 2026. We compare cost, latency, reasoning, and availability.
GPT-5Claude 4Gemini 2.5The rise of SLMs (small language models) and why they matter in LATAM
Why small models are eating a large slice of the market, especially in countries with higher infrastructure costs.
SLMOpen sourceLATAMEU AI Act and AI regulation in Mexico: what every SMB should know
Practical summary of the 2026 AI regulatory framework and what real obligations Mexican companies have.
RegulationAI ActComplianceRAG vs fine-tuning vs context caching in 2026: when to use each
Three techniques to make an LLM answer with your information. Which to pick based on case, budget, and volume.
RAGFine-tuningArchitectureComputer use and agents that control browsers: real cases in 2026
Agents that see the screen and move the mouse went from demo to production. Where they work, where they don't.
Computer useAgentsAutomationNative multimodality: voice, video, and image in business workflows
Real multimodal models are no longer a gimmick. B2B cases where voice, image, and video are unlocking new products in 2026.
MultimodalVoiceVisionAI costs in 2026: how token prices collapsed
Cost-per-token comparison between 2023 and 2026, what's driving it, and what it means for high-volume use cases.
CostsPricingStrategyMCP (Model Context Protocol) and the standardization of LLM tools
MCP is becoming the de facto standard to connect tools with LLMs. Why it matters, how it works, how we're using it.
MCPIntegrationsArchitecture