Efficient AI


Efficient AI Team is a collaboration between researchers from NVIDIA and MIT dedicated to the development and optimization of GPU-accelerated efficient AI computing. We focuses on pushing the boundaries of generative AI by designing models that are not only powerful but also efficient in terms of computational resources. We are committed to advancing the field of AI by making state-of-the-art models deployable, scalable and accessible.
Efficient AI

Latest News

3 Papers accepted by ICML 2025. 🎉🎉🎉
3 Papers accepted by ICML 2025. 🎉🎉🎉

Sana-1.5, Sparse VideoGen, and XAttention are accepted ICML 2025 🎉🎉🎉. See you at Vancouver Convention Center.

Latest Publications