한국어EN
Inference Infrastructure & ServingAMD

AMD launches Helios rack-scale system and Instinct MI455X

Made by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made

Launched the Helios rack-scale system and the Instinct MI455X

Why it matters

72 GPUs per rack, a single scale-up domain and 31TB of HBM4 — alternative silicon entering real-world deployment

Source — Primary sources first; secondary reports are labelled “Press”

Officialamd.comhttps://www.amd.com/en/blogs/2026/amd-launches-helios-the-highest-performing-rackscale-ai-infrastructure-solution.html

Added in Week 35, 2026 · See that week

On the map — ⑧ Inference Infrastructure line

  1. Previous stopGoogle Cloud previews its 8th-generation TPUs at Cloud Next
  2. This recordAMD launches Helios rack-scale system and Instinct MI455X
  3. Next stopDeepSeek-V4.1-Flash
See it on the map

Same week

imo-2026 project runs 7 frontier models on IMO 2026, publishes audit trailsOfficialUpstage releases Solar Open 2OfficialPlaintiffs file second amended complaint in Concord v. AnthropicPressSimon Willison blogs on OpenAI's accidental cyberattack against Hugging FaceOfficial

Nearby on the same axis

vLLM adds sharded weight-transfer engine built on Ray Direct TransportOfficialNVIDIA puts Groq 3 LPX into full productionOfficialResearchers release Random AttentionPaperTenstorrent releases the vLLM TT pluginOfficial