Themata.AI

AI is changing the world. Don't stay behind. Clear summaries, community insight, delivered without the noise. Subscribe to never miss a beat.

Privacy

Contact

Back to all news

claude anthropic compute-capacity ai-agents

Higher usage limits for Claude and a compute deal with SpaceX

anthropic.com

May 6, 2026

2 min read

🔥🔥🔥🔥🔥

68/100

Summary

A partnership with SpaceX will significantly enhance compute capacity for Claude. As a result, usage limits for Claude Code and the Claude API have been increased.

Key Takeaways

Anthropic has doubled the usage limits for Claude Code for Pro, Max, Team, and seat-based Enterprise plans.
A partnership with SpaceX will provide access to over 300 megawatts of new compute capacity, enhancing services for Claude Pro and Claude Max subscribers.
Anthropic has agreements for significant compute capacity expansion, including a 5 GW deal with Amazon and a $30 billion partnership with Microsoft and NVIDIA.
The company is expanding internationally to meet compliance and data residency requirements, focusing on partnerships with democratic countries.

Read original article

Community Sentiment

Mixed

Positives

Doubling the five-hour rate limits for Claude Code is a significant improvement, allowing users to maximize their productivity without hitting caps too quickly.
The removal of peak hour limits for Pro and Max accounts enhances accessibility, making the service more user-friendly for dedicated customers.
The partnership with SpaceX for orbital AI compute capacity indicates ambitious plans for scaling AI infrastructure, which could lead to innovative applications in the future.

Concerns

Some users view the doubling of the five-hour limits as a marketing stunt, suggesting it doesn't actually increase overall usage potential if weekly limits remain unchanged.
Concerns about the practicality of SpaceX's orbital compute plans raise doubts about their economic viability compared to terrestrial data centers.
There is skepticism regarding the effectiveness of Claude's models, with some users feeling that rate limits are irrelevant if the model's performance is lacking.