Anthropic taps SpaceX compute, boosts Opus API rate limits
▼ Summary
– Claude Code’s five-hour rate limits double across Pro, Max, Team, and Enterprise plans, and peak-hours throttling is removed for Pro and Max users.
– Anthropic signed an agreement to take all compute capacity at SpaceX’s Colossus 1 data centre, providing over 300 megawatts and 220,000 Nvidia GPUs.
– API rate limits for Claude Opus models have been raised considerably, with new ceilings published.
– Anthropic has existing compute agreements with Amazon, Google, Broadcom, Microsoft, Nvidia, and Fluidstack for additional capacity.
– The company will add inference capacity in Asia and Europe through its Amazon collaboration, focusing on democratic countries with strong legal frameworks.
Anthropic significantly boosted rate limits for Claude Code and the Claude Opus API on Tuesday, rolling out three immediate changes. The five-hour usage windows for Claude Code are now doubled across Pro, Max, Team, and seat-based Enterprise plans. Peak-hours throttling has been eliminated for Pro and Max accounts. API rate limits for Claude Opus models have also been raised substantially, with the updated ceilings now published.
The driving force behind these improvements is a new compute agreement with SpaceX. Anthropic has committed to taking the full compute capacity of SpaceX’s Colossus 1 data centre, which delivers over 300 megawatts of fresh capacity and more than 220,000 Nvidia GPUs within the month. This additional infrastructure will directly enhance service for Claude Pro and Claude Max subscribers, according to the company.
Anthropic also expressed interest in collaborating with SpaceX on developing multiple gigawatts of orbital AI compute capacity, though no formal agreement has been reached on that front.
The SpaceX deal adds to a growing list of compute commitments. Anthropic holds an up-to-5-gigawatt agreement with Amazon, including nearly 1 GW of new capacity online by the end of 2026. A 5 GW agreement with Google and Broadcom is set to begin coming online in 2027. The company also has a strategic partnership with Microsoft and Nvidia that includes $30 billion of Azure capacity, plus a $50 billion American AI infrastructure investment with Fluidstack.
Anthropic trains and serves Claude on a mix of hardware, including AWS Trainium, Google TPUs, and Nvidia GPUs. The company says it continues exploring opportunities to bring additional capacity online.
Some expansions will extend internationally. The collaboration with Amazon includes additional inference capacity in Asia and Europe, aimed at enterprise customers in regulated industries needing in-region infrastructure for compliance and data residency. Anthropic says it is deliberate about where it adds capacity, partnering only with democratic countries whose legal and regulatory frameworks can support investments of the relevant scale.
The company also reiterated an earlier commitment to cover any consumer electricity price increases caused by its US data centres. It is exploring extending that commitment to new jurisdictions as international expansion proceeds.
For users, the headline change is immediate. Pro and Max subscribers now have doubled five-hour usage windows for Claude Code and no peak-hours throttling. The Opus API rate limit increases also took effect Tuesday.
(Source: The Next Web)
