Googlebot IP Ranges
Google publishes the addresses of its robots in three files rather than one: Googlebot and the other common crawlers, special-case crawlers such as AdsBot, and user-triggered fetchers, which act when a person asks a Google product to fetch a page. Together they list 840 distinct IPv4 prefixes and 817 IPv6 prefixes. Unlike the AI crawlers on this site, Google needs no rented cloud: the sampled prefixes are announced by Google's own AS15169 and by AS396982, Google Cloud.
- Provider
- Googlebot
- Hosting network
- AS15169 (Google, host)
- Category
- Crawler
- Headquarters
- Mountain View, CA, USA
- Published IPv4 prefixes
- 840
- Registry
- Self-published list
Known IP ranges
Google does not announce these addresses under an ASN of its own. They are the prefixes it publishes in https://developers.google.com/static/search/apis/ipranges/googlebot.json, https://developers.google.com/static/search/apis/ipranges/special-crawlers.json, https://developers.google.com/static/search/apis/ipranges/user-triggered-fetchers.json, as fetched on 2026-10-16; the space itself is routed by AS15169 (Google). The list decides whether an address belongs to Googlebot - WHOIS for any of them returns the host network.
34.22.85.0/27
34.64.6.0/27
34.64.6.32/27
34.64.6.64/27
34.64.6.96/27
34.64.6.128/27
34.64.6.160/27
34.64.6.192/27
34.64.6.224/27
34.98.136.0/27
2001:4860:4801:2::/64
2001:4860:4801:c::/64
2001:4860:4801:f::/64 (IPv6)
What does a Googlebot IP mean in a privacy test?
A Googlebot address is Google reading a page, not a visitor. For a site owner it matters which of the three lists the address is in, because the three kinds of robot do not treat robots.txt the same way.
Three files, one schedule
The files are googlebot.json, special-crawlers.json and user-triggered-fetchers.json under developers.google.com/static/search/apis/ipranges/. All three carried a creationTime of 15 October 2026 when fetched on 16 October, a day apart, so Google regenerates them continuously and a cached copy goes stale quickly.
Google hands out its crawler space in small, regular blocks. Googlebot's own file has 170 IPv4 prefixes, 148 of them /27s and 22 /28s, plus 147 IPv6 /64s; most of the IPv4 sits in 66.249.0.0/16, the range Google has used for crawling for years. Special crawlers have 136 /27s, mostly in 74.125.0.0/16 and 66.249.0.0/16. User-triggered fetchers are the largest file at 534 /27s and 534 /64s, and most of those sit in 34.0.0.0/8 and 35.0.0.0/8, Google Cloud address space. In total the three cover 26,528 IPv4 addresses.
Which robot, and whether robots.txt applies
The file an address appears in tells you more than the user-agent does. Googlebot and the other common crawlers follow robots.txt. Special-case crawlers are Google's own name for robots whose behaviour depends on the product they serve, AdsBot being the familiar one. User-triggered fetchers run because a person asked for the page, and Google's documentation says they generally ignore robots.txt for that reason.
Google also documents a second check that needs no list: a reverse DNS lookup on a genuine crawler address returns a name under googlebot.com or google.com, and a forward lookup on that name returns the same address. A request claiming to be Googlebot that passes neither the list nor the DNS check is not Google.
Related tools
Frequently asked questions
What IP ranges does Googlebot use?
The 170 IPv4 and 147 IPv6 prefixes in googlebot.json, mostly /27s inside 66.249.0.0/16. Special crawlers and user-triggered fetchers have files of their own, 136 and 534 IPv4 prefixes respectively.
Why does a Googlebot address appear in my logs or a privacy test?
A Googlebot address is Google reading a page, not a visitor. For a site owner it matters which of the three lists the address is in, because the three kinds of robot do not treat robots.txt the same way.
How do I verify that a request really came from Googlebot?
Check the address against googlebot.json, or run a reverse DNS lookup: a real crawler address resolves to a name under googlebot.com or google.com, and that name resolves forward to the same address.
Why are some Google crawler addresses in 34.x and 35.x?
Those are Google Cloud addresses, announced by AS396982. Most of the 534 user-triggered fetcher prefixes sit there, while the classic crawler space in 66.249.0.0/16 is announced by AS15169.
Do Google's user-triggered fetchers obey robots.txt?
Generally not. They fetch a page because a person asked a Google product for it, and Google documents that they usually ignore robots.txt. Blocking them takes a firewall rule on the addresses in their own list.