Anthropic, the San Francisco-based AI company behind the Claude chatbot, has made a major deal with SpaceX to increase its computing power. The deal was officially announced on May 6, 2026.
Through this deal, Anthropic will get full access to SpaceX’s Colossus 1 data center in Memphis, Tennessee, with more than 300 megawatts of processing power starting this month. This comes as demand for Claude services continues to grow quickly, helping the company avoid server slowdowns.
The Colossus 1 facility is one of the biggest AI data centers in the world. Built by xAI in just 122 days, it became active in July 2024 and contains more than 220,000 Nvidia H100 GPUs. For Anthropic, using this center means better performance for users, especially subscribers of Claude Pro and Claude Max plans. The company said the extra capacity will improve service quality and reduce issues during busy hours when servers are overloaded.
Also Read: JioCarSync Launches in India to Bring Wireless Apple CarPlay and Android Auto Under Rs 2500
At the same time, Anthropic also announced major updates to its usage limits. The company doubled the five-hour rate limits for Claude Code on Pro, Max, Team, and seat-based Enterprise plans. It also removed peak-hour reductions that earlier slowed down services for Pro and Max users. Weekly quotas remain unchanged for now, but these updates solve many complaints from developers and heavy users about strict limits.
Anthropic also increased API limits for Claude Opus models. In the highest tier, input token limits increased from 2 million to 10 million tokens per minute, while output token limits increased from 400,000 to 800,000 tokens per minute. This allows larger projects and enterprise-level applications to run more smoothly.
The partnership with SpaceX is part of Anthropic’s bigger plan to strengthen and expand its AI infrastructure. The company already has a five-gigawatt deal with Amazon Web Services, with nearly one more gigawatt planned by the end of 2026 for Europe and Asia. Anthropic also has a five-gigawatt agreement with Google Cloud and Broadcom starting in 2027. In addition, it works with Microsoft Azure and NVIDIA through a $30 billion Azure resources partnership.
These partnerships show that Anthropic wants to spread its computing needs across different providers instead of depending on only one company. The company uses AWS Trainium chips, Google TPUs, and Nvidia GPUs to train and operate its Claude AI models.
Looking ahead, Anthropic also showed interest in working with SpaceX to build multiple gigawatts of AI computing systems in space. This could completely change how AI infrastructure works outside Earth. Financial details of the deal have not been shared yet.
Also Read: OnePlus Nord CE 6 Series Launched in India With Huge Batteries and 144Hz Displays
The timing is important for both companies. SpaceX, which is discussing a future IPO, gets a major AI customer. Anthropic, meanwhile, reduces its capacity problems while competing with OpenAI and Google DeepMind. Elon Musk, who earlier criticized Anthropic, is now indirectly supporting the company through SpaceX, even while xAI competes with products like Grok.
For users, the benefits will appear quickly. Paid Claude subscribers can expect fewer interruptions, faster API performance for coding and creative work, and more reliable services overall. Anthropic said its run-rate revenue crossed $30 billion in 2026, compared to $9 billion the previous year. The company also has more than 1,000 business customers spending over $1 million each year.