Anthropic doubles Claude Code session cap for Pro and Enterprise users
- Anthropic said on May 6 it doubled Claude Code’s five-hour usage limits for paid plans, including Pro, Max, Team, and seat-based Enterprise. (anthropic.com) - The concrete change is twofold: bigger five-hour budgets for Claude Code, plus the end of peak-hours limit reductions on Pro and Max. (anthropic.com) - It matters because Anthropic tied the increase to new compute capacity from a SpaceX deal, turning scarce GPU supply into product headroom. (anthropic.com)
Coding agents are only useful if they can stay on the job long enough to finish the annoying parts. That has been one of the biggest frustrations with C(anthropic.com)ile edits, and repeated test runs could slam into usage limits right when the session got interesting. Anthropic’s news is simple but imp(anthropic.com)ra headroom starts now. (anthropic.com) ### What changed? Anthropic said on May 6 that (anthropic.com) and seat-based Enterprise plans, with the change effective immediately. The company also removed the peak-hours limit reduction for Claude Code on Pro and Max accounts, so those users should see less daytime throttling on top of the bigger base allowance. (anthropic.com) ### Is this really a “session cap” change? Basically, yes — but the wording matters. Anthropic frames Claude Code access around rolling(anthropic.com)ut at exactly five hours. In practice, though, developers experience that system as a session budget: once the allowance is burned, the workflow stalls until the window resets. Doubling that budget means longer uninterrupted coding runs before hitting the wall. (anthropic.com) ### Who gets the upgrade? More than the he(anthropic.com)terprise, but Anthropic’s own announcement includes Max and Team too, plus seat-based Enterprise specifically. That matters because Claude Code has been moving from an individual power-user tool into a team workflow product, where rate limits hit harder when people are using it for real repository work instead of one-off prompts. (anthropic.com) ### Why did Anthropic do this now? The short answer is compute. A(anthropic.com)t access to the Colossus 1 data center, which outside reports describe as a massive GPU cluster built with Nvidia hardware. More capacity means Anthropic can loosen the product constraints that existed mostly because inference is expensive and scarce. (anthropic.com) ### Why does peak-hours throttling matter so much? Because it made the product feel least reliable when developers were mo(anthropic.com)us code review, asking Claude Code to trace a bug across files, or iterating on tests in the middle of the day, the budget could disappear faster. Removing that reduction on Pro and Max should make usage feel less random and more like something you can plan around. (anthropic.com) ### Does this fix the core limitation? Not entirely(anthropic.com)out usage in rolling windows rather than unlimited access. The company also kept the language pretty specific — it doubled the five-hour rate limits, not all limits everywhere forever. So the product is more usable for long tasks, but it has not turned into an always-on coding copilot with no ceilings. (anthropic.com) ### Why is this strategically important? Because coding agents compete on stamina as(anthropic.com)iles, run tests, and keep going is much more valuable than one that gives smart answers but runs out of runway mid-task. Anthropic is basically spending new compute supply to make Claude Code feel more like a dependable engineering tool and less like a demo with a timer attached. (anthropic.com) ### Bottom line? This is a capacity story disguised as a product tweak. Anthropi(anthropic.com)ode users can stay in flow longer. For developers, that is the difference between “helpful assistant” and “tool you can actually build with.” (anthropic.com)
