spacex opens its doors to anthropic on may 6, local time, spacex and the ai company anthropic jointly announced a significant agreement. under the terms of the deal, anthropic will be granted full access to the computational capacity of spacex’s colossus 1 data center.
this supercomputer is one of the largest and fastest-deployed ai computing facilities in the world, equipped with more than 220,000 nvidia gpus, including densely deployed h100 and h200 chips as well as the next-generation gb200 accelerators. leveraging this massive computational resource, anthropic plans to directly boost the service capacity for its claude pro and claude max subscribers and, effective immediately, comprehensively raise the usage caps for paying customers.
the specific enhancements include doubling the five-hour rate limit for claude code across the pro, max, team, and enterprise plans; eliminating the reduced rate limits on claude code during peak hours for pro and max accounts; and substantially increasing the api rate limits for the claude opus model. according to anthropic, these adjustments will significantly improve the interactive experience for high-frequency users. in addition, as part of the agreement, anthropic also disclosed its interest in collaborating with spacex and its partners to develop orbital ai computing capacity on the scale of several gigawatts, exploring the feasibility of deploying ai computing workloads on space-based platforms.
industry experts believe this move marks a deep integration between commercial spaceflight and cutting-edge ai computing infrastructure, potentially paving the way for a new paradigm in next-generation distributed intelligent computing. the world’s fastest ai supercomputer: claude users’ rate limits significantly increased
