AWS shows off rack-scale Trainium 3 custom silicon as the ultra server race heats up
BenBajarin · x · 2026-09-18
Analyst Ben Bajarin finally got a first-hand look at Amazon's fully custom rack-scale Trainium 3 system, originally shown at re:Invent last year. Per his chats with AWS staff, the race is now on around "ultra server" (scale-up domain) clusters — competing on how far vendors can scale within a single rack domain.
More from Infra
- HyperQwen seeks 4090/5090 owners to push Qwen local inference speeds — iamMess · 2026-09-18
- REBCO superconducting tape: 50-tesla fusion magnets and the rare-earth angle — Anen-o-me · 2026-09-18
- Where does VRAM go during LLM inference? Four buckets explained — blaizedsouza · 2026-09-18
- Planned US data centers hit 342GW, nearly 7x the 50GW installed today — BenBajarin · 2026-09-18
- $42 per billion input tokens with free output: an AI API price that looks like black magic — altryne · 2026-09-18
- openjev-sglang: Qwen3.6-35B-A3B serves Jev-compatible API, 64 tasks in under 1s — andersonbcdefg · 2026-09-18