PyTorch at NVIDIA GTC 2026: Join Us in San Jose!
PyTorch will have a strong presence at NVIDIA GTC 2026 in San Jose, offering demos, talks, hands-on labs, and opportunities to connect with PyTorch experts.
Why it matters
This event showcases the growing importance of PyTorch in the AI and machine learning landscape, with Meta and the broader community leveraging the framework for high-performance and distributed AI workloads.
Key Points
- 1Meta will showcase Helion, a PyTorch-native kernel authoring framework, and ExecuTorch for high-performance, Python-free, on-device inference
- 2Alban Desmaison, PyTorch core maintainer at Meta, will give a featured talk on how PyTorch powers high-performance AI
- 3The event will feature various PyTorch sessions and hands-on labs covering topics like optimizing PyTorch models, building fault-tolerant distributed AI training, and more
Details
The article announces that PyTorch will have a significant presence at the upcoming NVIDIA GTC 2026 conference in San Jose from March 16-19, 2026. At the Meta booth (Booth #338), attendees will be able to get hands-on experience with Helion, a new PyTorch-native kernel authoring framework, and see a demo of NVIDIA's Parakeet speech-to-text model deployed using ExecuTorch for high-performance, Python-free, on-device inference. Additionally, Alban Desmaison, a PyTorch core maintainer at Meta, will give a featured talk titled 'From Kernels to Clusters: How PyTorch Powers High-Performance AI', providing an overview of the PyTorch framework, its ecosystem, and insights into the technical roadmap and upcoming features. The event will also feature various other PyTorch-focused sessions and hands-on labs covering topics like optimizing PyTorch models, building fault-tolerant distributed AI training, and more.
No comments yet
Be the first to comment