The IEEE International Electron Devices Meeting (IEDM) is considered the premier forum where scientists and engineers come together to disclose, discuss, and debate the best recent R&D work in electron devices, semiconductors, and microelectronic technologies. The 71st annual IEEE IEDM conference will be held December 6-10, 2025, at the Hilton San Francisco Union.
Supporting the theme “100 Years of FETs: Shaping the Future of Device Innovations,” the 2025 IEDM technical program will consist of 295 presentations, plus a host of events.
“This year’s IEDM offers a preview of the next computing paradigm. We are witnessing the convergence of atomic-scale fabrication, 3D integration, and neuromorphic design, moving us beyond yesterday’s technology and even classical computing. From monolithic CFETs that redefine transistor density to in-memory computing that can mitigate the von Neumann bottleneck, more intelligent and more efficient technology is being built here,” said Gaudenzio Meneghesso, IEDM 2025 Publicity Chair, and Head of the Department of Information Engineering at the University of Padova.
“As AI is demanding unprecedented efficiency, the device community is rising to the challenge. IEDM 2025 showcases technologies that can power the next AI decade: from GaN chiplets for power delivery and silicon photonics for data transfer to 3D DRAM and FeFETs for in-memory computing. These device technologies will make the algorithms of tomorrow possible.”
“The pursuit of Moore’s Law has evolved from a simple race to shrink transistors into engineering in multiple dimensions. IEDM 2025 showcases this new era: we are scaling up with 3D stacking like CFETs and 3D DRAM, scaling laterally with atomic-layer-deposited channel materials, and scaling past CPUs with novel architectures like in-memory computing,” said John Paul Strachan, IEDM 2025 Publicity Co-Chair, Director of Peter Grünberg Institute, Forschungszentrum Jülich, and Professor at RWTH Aachen University. “This is the culmination of a century of FET innovation, pushing the boundaries of what is physically possible to compute.”
