AINewsnow

A 1T-parameter MoE reached 1,300 tokens per second per GPU on a public European supercomputer

๐Ÿ“… November 12, 2026 โ€ข domyn.com Domyn, the Milan company leading the EUROPA consortium, published a technical post describing training trials for a mixture-of-experts model on the JUPITER supercomputer. The same write-up appeared on the PyTorch blog the same day. According to the post, these are tโ€ฆ

Read the full story at DEV Community โ€” Machine Learning โ†—

Timeline ยท 1 report

  1. 2026-11-15 15:00 ยท DEV Community โ€” Machine Learning
    A 1T-parameter MoE reached 1,300 tokens per second per GPU on a public European supercomputer

More stories

  1. Jensen Huang resigns as Nvidia CEO amid calls for reinstatement and possible comeback โ€” Hard Fork (NYT)
  2. Researchers allege Chinese AI models were trained on export-banned Nvidia Blackwell chips โ€” Bloomberg AI
  3. Anthropic IPO oversubscribed nearly six times; can AI firm become new NVIDIA amid stock crash? Experts' views โ€” Mint AI
  4. Apple to unveil Siri AI smart glasses at WWDC, with Face ID coming to iPhone Duo 2 โ€” Tom's Guide AI
  5. A step toward defining the Standard Model on a lattice โ€” OpenAI News
  6. Bharat Semi raises $24M seed round to bring AI image sensors to market โ€” TechCrunch AI
  7. Grounded flights, Microsoft outage and AI turmoil fuel fears of wider tech breakdown โ€” Washington Post AI
  8. Earnings Calls Point to AI-Led Growth and Stronger Margins โ€” The Globe and Mail

Get the daily brief of stories like this at 6:30 every morning โ†’