Project
Open-source platform for managing the machine learning lifecycle, including experiment tracking, model packaging, and a model registry used across ML pipelines.
watchTowr says attackers are already pulling cloud credentials through MLflow's Tracking Server, and VulnCheck logged scanning against a FUXA path traversal a day later.
Perspective Coverage
5 publishersReality
The web tier's recovery target is 15 minutes; the on-prem ML pipeline's is 24 hours. A dev.to writeup covers the gap with a smaller model kept deployed in a second Azure region, and that stub runs continuously.
Reality
The richer sync carries training metrics, evaluation results and lineage into the Model Package version so a reviewer can validate from the registry alone, but none of it happens until someone sets AutoModelRegistrationEnabled.
Reality
The wrapper hashes code and output, sends only a digest out to be signed with RSA-PSS, and hands the reviewer a PDF. The cost of adopting it is a command prefix. The cost of believing it is trusting whoever holds the signing key.
Reality
The retailer benchmarked several time series foundation models over 101 rolling cutoffs of its own catalogue, and the one it shipped now covers both the 12-week and 52-week horizons that two separate models used to split.
Reality
One ModelTrainer and one ModelBuilder replace the SKLearn, PyTorch and XGBoost classes, with source directories synced at job launch. The container is now yours to supply.
Reality
A GitHub advisory describes an unauthenticated, full-read SSRF in the default MLflow tracking server that reaches the cloud metadata endpoint and reflects the response back to the caller.
Reality
CVE-2026-64849 lets anyone who can reach an MLflow tracking server make it fetch EC2 instance metadata and hand back the response. The fix is 3.15.0; the exposure is a default.
Reality