Anthropic and SpaceXAI sealed a computing agreement on May 6, 2026, granting the AI research firm access to the merged entity’s Colossus 1 supercomputer in Memphis, Tennessee. The announcement came during Anthropic’s annual developer conference in San Francisco, where executives highlighted the need for more horsepower to keep up with growing demand for its Claude AI models.

SpaceXAI, the product of a merger between Elon Musk’s SpaceX and his AI venture xAI earlier this year, is positioning the deal as a cornerstone of its upcoming public offering, slated for as soon as next month. By securing a high‑profile customer, the company hopes to demonstrate market traction for its ambitious plans, which include developing orbital AI compute capacity – essentially data centers in space.

The Colossus 1 system, built on a former Electrolux site in 2024, boasts roughly 220,000 Nvidia GPUs, ranging from H100 and H200 to the newer GB200 chips. SpaceXAI claims the supercomputer was assembled in just 122 days and now delivers over 300 megawatts of power to its customers. Anthropic will use this influx of compute to improve performance for Claude Pro and Claude Max subscribers, addressing recent complaints from developers about rate limits and service disruptions.

Developers have reported that the average user now spends at least 20 hours each week running Claude Code, a figure that underscores the strain on existing infrastructure. Anthropic’s access to the new capacity is expected to alleviate bottlenecks and expand the firm’s ability to serve enterprise clients.

The partnership arrives against a backdrop of environmental concerns. Residents near the Memphis data center have lodged complaints about emissions from gas turbines, and environmental activists recently protested at a SpaceX investor gathering ahead of the IPO.

Earlier this year, Musk publicly criticized Anthropic’s models, labeling their policies “misanthropic” and alleging bias. After a recent meeting with senior Anthropic staff, Musk posted on X that he was impressed with the company’s approach to ensuring its Claude system benefits humanity.

Anthropic’s cloud strategy has already involved massive commitments to major providers. The firm pledged $200 billion to Google’s AI cloud services and TPU chips, while also securing over $100 billion in Amazon cloud contracts for the next decade. Together with OpenAI, Anthropic now accounts for more than half of the $2 trillion backlog of contracts at the leading cloud providers.

With the new compute pipeline in place, Anthropic aims to solidify its position in the competitive AI landscape, while SpaceXAI hopes the partnership will reassure investors about the commercial viability of its supercomputing ambitions, both on Earth and beyond.

Used: News Factory APP - news discovery and automation - ChatGPT for Business