Abstract. Considerable progress has been made in the last year in the study of laser-driven inertial confinement fusion at the National Ignition Facility (NIF). Experiments have demonstrated symmetric capsule implosions with plasma conditions approaching those required for ignition. Improvements in computational models -in large part due to advances in non-LTE modeling -have resulted in simulations that match experimental results quite well for the X-ray drive, implosion symmetry and total wall emission [1].Non-LTE modeling is a key part of the NIF simulation effort, affecting several aspects of experimental design and diagnostics. The X-rays that drive the capsule arise from high-Z material ablated off the hohlraum wall. Current capsule designs avoid excessive preheat from high-energy X-rays by shielding the fuel with a mid-Z dopant, which affects the capsule dynamics. The dopant also mixes into the hot spot through hydrodynamic instabilities, providing diagnostic possibilities but potentially impacting the energy balance of the capsule [2]. Looking beyond the NIF, a proposed design for a fusion reactor chamber depends on lowdensity high-Z gas absorbing X-rays and particles to protect the first wall [3].These situations encompass a large range of temperatures, densities and spatial scales. They each emphasize different aspects of atomic physics and present a variety of challenges for non-LTE modeling. We discuss the relevant issues and summarize the current state of the modeling effort for these applications.