Perplexity limits feel reduced
Recent tracked changes that can explain it:
What’s going on
Nothing global changed on Perplexity — which is exactly what makes the squeeze confusing. Specific features were quietly cut while the rest of the product kept working. In February 2026 Deep Research on Pro dropped from a very high daily allowance to roughly 20 queries a month, Labs was halved from 50 to 25 a month, and file uploads went from effectively unlimited to 50 a week — all done silently and wrapped in soft language about 'allocating more compute per session,' so the change never announced itself and only surfaces when you lean on one particular feature. That per-feature pattern is the thing to read: Deep Research stalling far sooner than it used to while ordinary Pro search feels normal is a specific quota cut, not an account problem or an outage. Because the limits live per feature, so does the fix — check each one's remaining quota in settings rather than assuming a single number covers everything, batch your Deep Research questions instead of firing them off exploratively, and spread Labs runs across the month rather than clustering them. On a promo-code Pro account, a separate May 2026 change throttled those accounts to around 100 advanced queries a week as part of a resale crackdown — a policy action against how the account was obtained, not a general quota. And when your work genuinely depends on high-volume research, the Max tier at $200 removes the Labs and Research caps entirely, which is the structural answer once pacing isn't enough. Just don't let a comfortable month of ordinary search convince you the Deep Research tank is still full; that quota drains on its own schedule.
Perplexity has become more restrictive over time, mainly in usage limits and quotas.
Some recent constraint changes — monitor before depending on it.
Related problems
- You hit your Claude usage limit faster than expected →
- Cursor usage runs out or costs more than before →
- GitHub Copilot costs more or runs out of credits →
- DeepSeek API costs or models changed under you →
- An ElevenLabs voice or model you relied on stopped working →
- Tabnine removed its free plan and raised prices →