Hi
Forwarding this message from wikitech-l as it's also relevant to this list. The the new API rate limits https://www.mediawiki.org/wiki/Wikimedia_APIs/Rate_limits were deployed as planned yesterday.
If you encounter any issues with these rate limits, please let us know. Edge cases that are known to be potentially problematic include bots that use owner-only tokens to authenticate but do not support cookies, gadgets that make heavy use of the API, and external browser-based apps or games that make a lot of API calls.
We have done our best to mitigate these and provide options, but please do contact us at bot-traffic@wikimedia.org if you are unsure as to the best path forward.
Best Jonathan
---------- Forwarded message --------- From: Jonathan Tweed jtweed@wikimedia.org Date: Fri, 24 Apr 2026 at 10:36 Subject: Roll out of global API rate limits for identified requests To: Wikitech-l wikitech-l@lists.wikimedia.org
Hi
We are now ready to start rolling out the next phase of global API rate limits. As previously announced https://lists.wikimedia.org/hyperkitty/list/wikitech-l@lists.wikimedia.org/thread/GBFZTN3A233IR6F4HEENCIUCVI2ZH6YB/, this will apply to all API requests, including those that we class as identified.
The full set of limits https://www.mediawiki.org/wiki/Wikimedia_APIs/Rate_limits have now been published on wiki, along with answers to frequently asked questions https://www.mediawiki.org/wiki/Wikimedia_APIs/Rate_limits/FAQ that we received after the roll out of limits to anonymous requests. We will gradually ramp down to the published limits, starting from next Tuesday 28th April. We expect to be at the final limits by the end of May.
As part of this change we have moved from per-hour limits to per-minute limits, so please bear that in mind when you first look at the docs! This is preferable, as it helps smooth aggregate request rates and also minimises the impact on any user that inadvertently hits a limit.
These limits are based on extensive analysis of API request data, but it is possible that some bots or gadgets may still trigger rate limits. We will be actively monitoring all relevant spaces, so please do let us know if you need help to work out the best way forward should this happen.
Thank you for your engagement with this. I hope from our actions it is clear that we are not trying to limit community usage, but protect our projects from automated usage at a scale we cannot support https://diff.wikimedia.org/2026/03/26/quo-vadis-crawlers-progress-and-whats-next-on-safeguarding-our-infrastructure/ without ensuring it is a fair and sustainable use https://www.mediawiki.org/wiki/MediaWiki_Product_Insights/Responsible_Reuse of resources.
Best Jonathan