Perplexity seems to have quietly reduced usage limits for Pro users [U: Official explanation]

Update 19/05/26 – 04:53 pm (IST): Following widespread user complaints, a Perplexity spokesperson has issued a statement to Android Authority clarifying the situation. The company attributes the varying usage limits to a crackdown on fraud and the unauthorized resale of promotional codes, rather than a stealthy system-wide downgrade for standard subscribers.

Here’s the statement shared by the publication:

Some users are seeing different usage limits because their accounts are tied to promotional-code offers, and we’ve had to adjust enforcement due to fraud and unauthorized resale of those codes, including cases where people may have unknowingly purchased invalid codes from third parties. We’re working to make the applicable limits clearer for affected users and will continue helping anyone who believes they were impacted in error, while continuing to enforce our terms against abuse.

Original article published on May 15, 2026, follows:

Many Perplexity Pro users are complaining about lowered limits over the past few days. The company might have quietly changed the query cutoff, leading to much worse rate limits.

There are various posts on X and threads on Reddit talking about the issue. Casual usage with 10-20 queries per day exhausts the weekly quota, which is mediocre for a subscription service. Weekly attachment uploads are hitting a wall as well. You can find more reports here and here.

It’s worth noting that the new query restrictions only apply to advanced AI models, which are exclusive to the Perplexity Pro subscription, such as Gemini 3.1 Pro Thinking. Regular models remain unaffected.

With similar usage as before, some users report hitting weekly token queries way faster than expected, despite no changes to the volume of prompts.

We’ve experienced something similar as well. A team member mentioned that they got hit with the usage limit notice for advanced models, despite lower-than-average usage this week.

Users are assuming that Perplexity might have changed the token limits without directly informing users. A comment on Reddit claims that the “remaining tokens” have been changed from 200 to 100, and a post on X says that the queries per week have been reduced to 100/150. However, both claims lack strong sources, so this is mere speculation.

Some were contemplating switching over to Grok because of this, but xAI has also throttled video, image, and voice features. You can read about that here.

As expected, people are fairly annoyed since these usage caps directly contradict marketing claims. It’s supposed to offer the “best” token caps for “most” users, but the company is clearly attempting to push the “Max” subscription now.

The thresholds for Perplexity Pro are not transparent in the first place, as some posts point out. There are accusations of aggressive query cutdowns, about which paying customers are seemingly kept out of the loop.

While other companies, such as Google, have also introduced such query limitations before, representatives acknowledged this after a while. However, no statement from Perplexity exists at the time of writing, both on their socials and blog. Clearly specified quotas and straightforward communication to users would’ve avoided the entire fiasco.

We’ll update the article in case Perplexity issues a statement about the changes. In the meantime, users can just select the “Best” option from the list of models, and it’ll automatically allocate a suitable model for the query. Either way, these new limits feel like a bait-and-switch tactic.

Disclaimer: The Featured Image was generated with AI.

We stand out from the tech-media crowd because we break news stories; we mainly bring you stuff that you won’t find anywhere in the mainstream tech media. Our stories have been picked up by some of the world’s most popular websites and media outlets—more info is available here.