Anthropic announced major infrastructure expansions on May 6, 2026, doubling Claude Code's five-hour rate limits across Pro, Max, Team, and Enterprise plans while signing a landmark compute partnership with SpaceX. The company will gain access to all compute capacity at SpaceX's Colossus 1 data center in Memphis, Tennessee, providing over 300 megawatts of new capacity—equivalent to more than 220,000 NVIDIA GPUs.
Immediate Capacity Improvements for Claude Users
The usage limit increases take effect immediately across Anthropic's product line:
- Claude Code's five-hour rate limits doubled for Pro, Max, Team, and seat-based Enterprise plans
- Peak hours limit reduction removed for Pro and Max accounts
- API rate limits for Claude Opus models substantially raised
- New capacity benefits Claude Pro and Max subscribers directly
The SpaceX partnership delivers access to H100, H200, and GB200 accelerators within the month, representing one of the largest single compute deals in AI history.
SpaceX Partnership Opens Door to Orbital Computing
Beyond the immediate Colossus 1 access, Anthropic expressed interest in developing "multiple gigawatts of orbital AI compute capacity" with SpaceX. This ambitious vision suggests future AI training and inference could occur in space-based data centers.
The SpaceX agreement complements Anthropic's existing infrastructure portfolio:
- Amazon: Up to 5 GW total capacity, with nearly 1 GW coming online by end of 2026
- Google & Broadcom: 5 GW capacity scheduled for 2027
- Microsoft & NVIDIA: $30 billion Azure capacity agreement
- Fluidstack: $50 billion investment in American AI infrastructure
Strategic Implications for AI Infrastructure
The 300+ megawatt Colossus 1 facility represents immediate, production-ready capacity rather than future promises. With over 220,000 NVIDIA GPUs available, Anthropic gains substantial competitive advantage in training larger models and serving more concurrent users.
The announcement generated significant community discussion, receiving 373 points and 314 comments on Hacker News. Industry coverage spanned Bloomberg, CNBC, and specialized publications, reflecting the deal's importance to AI infrastructure development.
Key Takeaways
- Anthropic doubled Claude Code rate limits and removed peak-hour restrictions for Pro and Max users
- SpaceX partnership provides 300+ megawatts of compute capacity (220,000+ NVIDIA GPUs) within one month
- The Colossus 1 facility includes H100, H200, and GB200 accelerators for immediate deployment
- Anthropic now has partnerships totaling over 15 GW of compute capacity across Amazon, Google, Microsoft, SpaceX, and Fluidstack
- Future collaboration may extend to "multiple gigawatts of orbital AI compute capacity"