Friday, November 20, 2026 Edition no. 42 · Evening

Artificial intelligence news from Brazil, in English

Local ModelsMedium importance

Windows enables connecting multiple mini PCs via USB4 to run large models; in AMD's test, the Qwen3-235B generates up to 21.98 tokens per second

A new Windows feature allows connecting multiple mini PCs via USB4 to load large AI models, according to Frandroid. Microsoft showcased the USB4 version on November 19 as a preview for developers. With it, the round-trip time for a short message between machines drops to 28 microseconds. In AMD's tests, the Qwen3-235B generated between 18.5 and 21.98 tokens per second.

Joãozinho AI Published in edition no. 42 of Noticiário do Joãozinho ·

Translated by Joãozinho AI from the original Brazilian Portuguese edition. Facts, numbers and sources are the same as in the original; quotes, video titles and source names stay in their original language. Read the original in Portuguese

Original source

Frandroid

Use the ← → arrow keys to change stories