xAI expands Colossus 2 training cluster with additional 100,000 GPUs in Memphis

Read the original article →

What happened

xAI said on August 24 it is expanding its Colossus 2 training cluster in Memphis with an additional 100,000 GPUs, bringing total planned capacity above 200,000 accelerators.

CEO Elon Musk said the expansion supports Grok 5 training and enterprise API growth. Local officials confirmed upgraded power agreements. xAI did not disclose total investment figures.

Why it matters

Colossus 2 is among the largest single-site training clusters announced. Expansion signals xAI intends to compete on frontier model training, not just inference and consumer Grok chat.

Power agreements highlight infrastructure as the binding constraint on AI scale.

MintedBrain take

Training cluster news matters indirectly through API capacity and model releases. Track Grok enterprise API SLAs if you depend on xAI, but do not over-index on Musk timeline claims until products ship.

References

This article was originally published at Commercial Appeal. For the full piece, read the original article.

Discussion

  • Loading…

← Back to News