Anthrophic has announced a new partnership to increase the computing power behind its Claude AI models, as demand for AI tools continues to grow rapidly.
The company said the agreement will “substantially increase” its compute capacity, helping it support more users across Claude Code and the Claude API. Following the deal, Anthropic has already increased usage limits for developers and paid users.
The partnership is centered around SpaceX’s massive AI infrastructure system known as Colossus 1, a large-scale AI data center powered by more than 220,000 Nvidia GPUs. Reports say the system provides over 300 megawatts of compute capacity, making it one of the largest AI computing environments currently available.
As part of the update, Anthropic said it has doubled Claude Code’s five-hour usage limits for Pro, Team, Max, and Enterprise users. The company also removed some peak-hour restrictions and increased API limits for Claude Opus, allowing developers to run larger and more complex workloads.
This matters because AI companies are now facing a major infrastructure challenge. Training and running advanced AI models requires enormous computing power, especially as users begin using AI agents for coding, research, automation, and long-running tasks.
Claude Code in particular has become popular among developers because it can write code, edit files, execute commands, and complete multi-step tasks. However, growing usage has also placed pressure on Anthropic’s infrastructure, leading to limits on how much users could do within a session.
The new compute deal is expected to reduce some of those limits and improve reliability across Anthropic’s services.
What makes the partnership more interesting is the relationship between Anthropic and . Musk has publicly criticized Anthropic multiple times over the past year. He has accused the company of abandoning AI safety principles and claimed that some leading AI firms are becoming too politically biased and commercially driven.
Musk has also repeatedly attacked rival AI companies while promoting his own startup, xAI. In several interviews and social media posts, he grouped Anthropic together with OpenAI as companies he believes are not being fully transparent about AI development.
However, despite that public criticism, Anthropic is now working directly with SpaceX infrastructure. This creates an unusual situation where competition and partnership exist at the same time. It also shows how intense the race for compute power has become in the AI industry. Companies may disagree publicly, but access to infrastructure is becoming too important to ignore.
Furthermore, the agreement highlights how AI companies are increasingly competing for GPUs, energy, and data centers, not just better models. Compute power has become one of the most valuable resources in artificial intelligence.
Anthropic has recently signed several infrastructure agreements, including partnerships with Google, CoreWeave, and Broadcom. The SpaceX deal adds another major layer to that expansion strategy.
There is also a longer-term ambition behind the partnership. Reports suggest Anthropic and SpaceX are exploring the possibility of building AI compute infrastructure in space, using orbital systems powered by solar energy. While still early, the idea reflects how companies are beginning to think beyond traditional data centers as AI demand keeps increasing.
The bigger picture is becoming clearer across the AI industry. The race is no longer only about building smarter AI models. It is now equally about who controls the infrastructure needed to run them at scale.
The SpaceX partnership gives Anthropic more room to grow Claude’s capabilities. For the wider industry, it shows that even fierce rivals may end up working together when the demand for AI infrastructure becomes too large to handle alone.
Read also: Apple Plans to Let Users Choose AI Models in iOS 27