Databricks MMF, Genie, and Amazon Quick enable autonomous replenishment by letting forecasts trigger routine reorders and involving planners only in exceptions, according to a post by AWS and Databricks architects. The post states that replenishment decisions repeat every planning cycle for every item in the inventory catalog, and in most retail organizations planners still make them by hand—exporting the forecast, checking each stock keeping unit (SKU) against what suppliers can ship, and placing orders one line at a time. It notes that pre-trained models now forecast demand across a full catalog with little tuning, but the manual work between forecast and order does not scale on a catalog refreshed every day. The post cites IHL Group’s 2025 Inventory Distortion Study putting the combined cost at about $1.73 trillion a year, roughly 6.5 percent of global retail sales, and IHL’s 2026 study reporting about $1.7 trillion, or roughly 5 percent of global retail revenue, with out-of-stocks accounting for approximately two-thirds of the total and overstocks the remaining third. It describes routine decisions as following a fixed rule—demand rises, a supplier has stock, the order is placed—while exceptions include when no supplier can cover demand or spend crosses an approval threshold. Databricks produces the forecast; the Databricks Many Model Forecasting (MMF) solution accelerator runs Chronos-2, an open-source, pre-trained forecasting model from