AI Limits & Quotas Reference

Maintained reference

Limits and reset windows for 16 widely used AI tools.

Message caps, context windows, API rate limits and reset windows — evidence-tiered, dated, and linked to primary sources. When a vendor does not publish a number, we say so instead of guessing.

16 tools 85 limits tracked oldest check 2 Aug 2026 4 need recheck

Understand the limit you reached

Five practical guides explain where to check, what resets, and when an upgrade might duplicate something you already have.

Quota reset forecaster

Check documented reset boundaries. Your account notice takes priority.

Tool

Limit you hit

Reference date and time

— time to reset

window starts — next boundary —

The finding

Vendors publish the number when the number is what you buy

Sort the surfaces by how completely their limits are documented and you get a clean gradient, not a muddle. Midjourney sells GPU hours by the hour and documents every plan to the decimal. The APIs sell tokens and publish their tiers in full. Consumer assistants sell a flat monthly subscription — and there the limit stops being the product and becomes a cost control, which is a thing companies describe rather than quantify. Google replaced Gemini’s daily prompt counts with undisclosed “compute” units; Cursor calls its allowance “generous included usage.” Same companies, different disclosure, depending on what sits on the invoice.

● Documented The vendor publishes the exact figure on a public page. Quote it and cite it.

◐ Console-only The number exists and is authoritative, but you must log in to your own account to see it. Not quotable in the abstract.

○ Observed Reported by the vendor in a changelog, blog or support reply, but absent from the reference docs. Reliable, but it can move silently.

⨯ Undisclosed The vendor has chosen not to publish a figure. Any hard number you find elsewhere is someone’s guess.

The reference

Every tracked limit

Swipe or scroll horizontally to see every column.

Tool	Plan	Limit	Window	When exhausted	Evidence

Methodology

How this page is maintained

Primary sources only. Every figure is taken from the vendor’s own documentation, support centre, changelog or pricing page. Secondary reporting is used to find a claim, never to publish one.
Evidence tier is recorded per limit, not per tool. A single vendor routinely publishes its API limits in full while declining to publish its consumer message caps. Marking the tool as a whole would hide that.
Undisclosed stays undisclosed. Where a vendor has stopped publishing a number, no figure is substituted from a third-party estimate. The gap is the finding.
Verification is evaluated per limit. Version-two claims carry their own dates; legacy rows temporarily inherit the dataset date while they are migrated. A figure not re-checked inside 30 days is marked for recheck.
Superseded figures are kept, not deleted. Readers arrive holding the old number and need to see that it changed.
A tool we could not verify is held out, and named.

Budget the work before you start

Compare generation attempts, review time and upfront spending using your own assumptions.