Higher Limits for Claude Code and API
Anthropic just dropped significant usage limit increases for Claude Code and the Claude API. Effective immediately, here's what changed:
- Claude Code rate limits doubled for Pro, Max, Team, and seat-based Enterprise plans. The five-hour window now allows twice the requests.
- Peak hours limit reduction removed for Pro and Max accounts. No more throttling during busy times.
- API rate limits raised considerably for Claude Opus models (specific numbers in the source table).
These moves directly address developer complaints about hitting ceilings during heavy coding sessions. If you've been frustrated by Claude Code cutting you off mid-refactor, that's over.
SpaceX Compute: 300+ MW of GPUs
Anthropic signed an agreement with SpaceX to use the entire compute capacity of their Colossus 1 data center. That's over 300 megawatts of capacity — more than 220,000 NVIDIA GPUs — coming online within the month. This capacity is earmarked for Claude Pro and Max subscribers.
This isn't Anthropic's only compute deal. They've been stacking infrastructure like crazy:
- Up to 5 GW with Amazon — nearly 1 GW by end of 2026.
- 5 GW with Google and Broadcom — coming online in 2027.
- $30 billion of Azure capacity via Microsoft and NVIDIA.
- $50 billion investment in American AI infrastructure with Fluidstack.
They're also exploring orbital AI compute with SpaceX — multiple gigawatts in space.
International Expansion and Community Commitments
Anthropic is expanding internationally, particularly for regulated industries (finance, healthcare, government) that need in-region infrastructure for compliance. Some Amazon capacity will go to Asia and Europe.
They're being selective about where they build — partnering with democratic countries with secure supply chains for hardware, networking, and facilities.
And they've committed to covering any consumer electricity price increases caused by their US data centers. They plan to extend that commitment internationally, partnering with local leaders to invest back into host communities.
What This Means for Developers
If you use Claude Code or the API, today's limit increases are immediately beneficial. The SpaceX deal signals that Anthropic is serious about scaling — not just for training, but for inference capacity that directly impacts user experience. With multiple hardware partners (AWS Trainium, Google TPUs, NVIDIA GPUs), they're avoiding vendor lock-in.
The international expansion also means better latency for developers outside the US, especially in regulated industries where data residency is mandatory.
Next Steps
Update your Claude Code workflows — the higher limits mean you can run more aggressive automation or longer debugging sessions without hitting rate limits. For API users, check the new Opus rate tables and adjust your batching accordingly. If you're in a regulated industry, keep an eye on Anthropic's international rollout for potential latency improvements.
