Amazon Launches AI Supercomputer with Anthropic, Boosts AWS AI Race

What happened

Amazon has completed a major AI supercomputer project, Project Rainier, in partnership with Anthropic, a leading AI research company. Anthropic will double its investment in Amazon Web Services’ AI infrastructure and begin scaling its Claude AI models on this new system, which is powered by nearly 500,000 Trainium 2 chips.

Why it matters

• The deal strengthens Amazon’s position in the AI infrastructure race, providing predictable, long-term demand from a top AI company.
• Amazon’s in-house Trainium 2 chips reduce its reliance on Nvidia and allow for more optimized, cost-effective AI model training.
• The partnership signals Amazon’s ambition to be a full-stack AI provider, from hardware to cloud services, challenging competitors like Google and Nvidia.

Key facts

1. 1. Project Rainier features nearly 500,000 Trainium 2 chips, making it one of the world’s largest AI compute clusters.
2. 2. Anthropic is expected to use more than 1 million Trainium 2 chips for Claude model workloads by the end of 2025.
3. 3. Amazon has committed $8 billion to Anthropic, while Google has invested $3 billion in the company.