PyTorch and TorchTitan used to train 1T-parameter MoE on EU public infrastructure

A PyTorch blog post describes training a roughly 1-trillion-parameter Mixture-of-Experts model on shared EU public infrastructure using PyTorch and TorchTitan, a task previously deemed unfeasible.

Read the full story at PyTorch Blog ↗

Timeline · 1 report

2026-11-12 12:00 · PyTorch Blog
Training a 1T-Parameter MoE on EU public infrastructure with PyTorch and TorchTitan

More stories

PyTorch Models Can Now Run on GPU in the Browser via ExecuTorch — PyTorch Blog
The phantom life of Alice Huang, alleged Chinese spy who had affair with Nvidia CEO — CSET Georgetown
Google DeepMind unveils AlphaGenome 2 for cell-type-specific DNA variant predictions — Google DeepMind Blog
Google DeepMind's AlphaProof Code formally verifies Signal's cryptographic code in four weeks — Google DeepMind Blog
Google launches Gemini Decision 1 model and Decisions API for fast closed-set choices — Google Cloud AI Blog
Google launches Tabletop, an experimental AI platform for group planning and collaboration — Google AI Blog
United Airlines to Ban AI Bots From Booking Flights — Wall Street Journal Technology
Google expands Gemini 4 Carbon access to Pro plan users in AI Mode — Google DeepMind Blog

Get the daily brief of stories like this at 6:30 every morning →