The Time Series Optimized Transformer Setting New Standards in Observability

Datadog introduces Toto, a groundbreaking foundation model for time series forecasting, trained on over one trillion data points. Optimized specifically for observability metrics, Toto delivers state-of-the-art zero-shot performance across multiple domains. Its novel architecture includes factorized space-time attention and a Student-T mixture model, enabling more efficient, accurate, and scalable predictions for complex, high-frequency infrastructure data. Toto marks a major step forward in real-time system monitoring and predictive analytics.