Datadog AI Research Lab Launches First-Ever Observability Foundation Model

Datadog, Inc. (NASDAQ:DDOG) unveiled on May 21 the first releases from its newly established Datadog AI Research division. The company introduced an open-source foundation model called Toto and an observability benchmark named BOOM. This launch coincides with the company’s advancement toward FedRAMP High authorization to serve mission-critical federal workloads.

A closeup of a laboratory technician monitoring a computer for the Cloudbreak platform.

Toto is the first open-source foundation model designed explicitly for observability. It is trained exclusively on Datadog’s internal telemetry metrics. The model achieves “state-of-the-art performance” compared to existing time series foundation models (TSFM), enabling instant anomaly detection and capacity planning without per-series tuning. This capability becomes critical when monitoring billions of ephemeral time series that characterize modern cloud environments.

Alongside Toto, Datadog introduced BOOM, the largest public benchmark for observability metrics. It provides 350 million observations across 2,807 real-world multivariate series. The benchmark captures the unique challenges of production telemetry, including scale, sparsity, spikes, and cold-start issues.

The AI research initiative supports the company’s broader expansion into government markets. Datadog announced it has achieved Federal Risk and Authorization Management Program (FedRAMP) High authorization. This authorization allows the company to serve federal agencies with stringent security requirements.

Datadog, Inc. is a cloud monitoring and security company. It provides observability and analytics for IT infrastructure, applications, and security. Its platform includes infrastructure monitoring, application performance monitoring, log management, cloud security, and incident management.

READ NEXT: 20 Best AI Stocks To Buy Now and 30 Best Stocks to Buy Now According to Billionaires.