Claude Code Rate Limits Just Doubled — Here's Why

If you've been hitting Claude Code's rate limits mid-session — watching your task queue stall at the worst possible moment — something changed on May 6. Anthropic doubled the limits. No gradual rollout, no waitlist. Immediate.

The reason: Elon Musk's data center.

What Anthropic Did

Three changes took effect on May 6, 2026: Claude Code's five-hour rate limit doubled for Pro, Max, Team, and seat-based Enterprise plans. Peak-hour limit reductions for Pro and Max Claude Code accounts were removed. And Claude Opus API rate limits were significantly increased.

You don't need to do anything to receive these changes. They're already live.

The SpaceX Deal Behind It

Anthropic signed an agreement with SpaceX to use the full compute capacity of SpaceX's Colossus 1 data center — over 300 megawatts, corresponding to more than 220,000 NVIDIA GPUs — available to Anthropic within the month.

Colossus 1 is the Memphis supercomputer that xAI (now absorbed into SpaceX) originally built to train Grok. Not a partial lease — the entire data center.

The Musk angle is hard to ignore. Anthropic partnering with SpaceX may seem like an unlikely move — Musk recently called Anthropic "misanthropic and evil" and has long been critical of CEO Dario Amodei. But in a post on X after the deal, he said he was "impressed" with the Anthropic team and that Claude will "probably" be good. Weeks after calling them the most hypocritical company in AI, he handed them his entire data center.

The most plausible read: this is an "enemy of my enemy" play — Musk's lawsuit with Sam Altman is ongoing, and anything that helps Anthropic pull ahead of OpenAI is, from Elon's perspective, a win.

Why the Limits Were Broken in the First Place

Dario Amodei explained the root cause at Anthropic's developer event: "We planned for a world of 10x growth per year. In Q1 2026, we saw 80x annualized growth per year in revenue and usage."

That's not a rounding error. The infrastructure simply wasn't built for that trajectory. Colossus 1 fixes the supply side of that equation immediately — it's online and running now, not in six months.

The API Numbers That Actually Matter

If you're building on the Claude API, here's the concrete change. For Tier 1 users, the maximum input tokens per minute jumped from 30,000 to 500,000. Maximum output tokens per minute increased from 8,000 to 80,000.

Tier 1 input tokens went from 30K to 500K — a 16x increase. Tier 4 went from 2M to 10M. Output token throughput increased between 2x and 10x depending on your tier.

These are not incremental bumps. For Tier 1 users especially, this is the difference between a tool you can run extended agentic workflows on and one you're constantly babysitting.

The Bigger Infrastructure Picture

The SpaceX deal is one piece of a much larger compute buildout. Anthropic's infrastructure agreements now include up to 5GW with Amazon (nearly 1GW online by end of 2026), 5GW with Google and Broadcom (coming in 2027), a $30 billion Azure partnership with Microsoft and NVIDIA, and a $50 billion US infrastructure investment with Fluidstack.

Anthropic has also expressed interest in partnering with SpaceX to develop multiple gigawatts of orbital data center capacity — satellites running AI inference in low Earth orbit. That's the long-term direction. Today's limit increase is the immediate one.

What This Means If You Use Claude Code

If you're on Pro or Max, your five-hour session bucket is now twice as large and no longer throttled at peak hours. Long refactors, deep repo analysis, multi-step agent runs — all of these hit the wall less.

If you're on the API building agents or automations, the headroom increase is substantial. Tier 1 is especially improved — 16x input token throughput is the kind of change that makes previously impractical workloads viable.

Nothing to update. Nothing to enable. Just more room to work.

Sources: Anthropic, The New Stack, Engadget, MindStudio