In summary:
- PCWorld reports that Anthropic is doubling Claude AI usage limits during off-peak hours (weekdays 2 p.m. to 8 a.m. ET) until March 27, 2026.
- This promotion covers all Claude models including Opus, Sonnet, and Haiku across subscription tiers, with extra usage not counting against weekly limits.
- The extended limits provide users an ideal opportunity to extensively test powerful frontier AI models like Claude Opus 4.6 for complex tasks.
Haven’t ventured from ChatGPT yet? One of my best tips for beginning AI users is to switch models and providers every so often, and Claude developer Anthropic is serving up a golden opportunity to do just that.
From now until March 27, Anthropic is doubling the five-hour usage limit for all Claude models across every subscription tier except Enterprise. The increased limits will work on the Claude web interface as well as the Claude Desktop app, Claude Code, and Claude Cowork.
There is a catch: The doubled usage limits apply only during off-peak hours, meaning you’ll see the boosted limits on weekdays from 2 p.m. to 8 a.m. Eastern time or 11 a.m. to 5 a.m. Pacific time.
Specifically, the extra usage limits apply to Claude’s five-hour usage windows, which begin after your first prompt; if you burn through your usage cap during that window, you’ll be timed out until the clock runs out, and then your next five-hour window begins following your subsequent prompt.
Claude users also have an overall weekly usage limit, but your extra usage during this special promo period won’t count against your weekly limit, Anthropic says.
Anthropic’s Claude deal makes for the perfect time to try a new AI provider for a while, and it’s also a great way to check out a “frontier” AI model if you haven’t done so already.
Anthropic’s current high-end model for consumer use is Claude Opus 4.6, which was released early last month. Opus 4.6 boasts such cutting-edge features as “adaptive” thinking, which allows the model to adjust how long to ponder its answers based on the context of the conversation–the tougher the problem, the longer it’ll think about it.
A step down from the high-end Opus model is Sonnet 4.6, another thinking model that’s better suited for everyday tasks like crunching numbers in Excel or other office tasks. Then there’s Haiku 4.5, a lightweight model that’s ideal for straightforward tasks like proofreading, summarizing news articles, and making grocery lists based on recipe ideas.
Generally speaking, even free users will get plenty of Haiku prompts answered during a decent chat session, while you’ll hit your Sonnet usage cap more quickly. Finally, you might get cut off from Opus after just one or two prompts.
That’s why it’s a great time to try Opus 4.6 during Anthropic’s off-peak deal, particularly if you’ve never tried an AI model this powerful before.
Give Opus a task it could really chew on–such as, say, evaluating your resume and identifying areas of weakness or opportunities for a career pivot–and see what happens.
Pro tip: Ask Sonnet 4.6 to generate a detailed prompt for a tough Opus task, and then hand over the prompt to Opus.
Author: Ben Patterson, Senior Writer, PCWorld
Ben has been writing about consumer technology for more than 20 years, and now focuses his reporting on AI as it relates to the basic human experience. His coverage of artificial intelligence interrogates the latest LLMs, and how they can be used at work and at home to be best prepared for the AI revolution. “AI is going to change our lives sooner than we think,” Ben writes. “Our best way to adapt is by using it every day.” Ben has been a PCWorld author since 2014, and has covered everything from laptops to security cameras before launching PCWorld’s AI beat. Ben's articles have also appeared in PC Magazine, TIME, Wired, CNET, Men's Fitness, Mobile Magazine, and more. Ben holds a master's degree in English literature.