Baidu · Search index
Baiduspider
Reverse DNS is not a waypoint on the road to an IP list here: it is where verification stops, permanently, because Baidu has twice refused in writing to publish one:「Baiduspider的IP池是不断变动的，我们无法提供IP全集」("Baiduspider's IP pool changes constantly; we cannot provide the complete set of IPs"), and three years later「IP地址范围动态变化不固定，我们无法对外公布」("the IP address ranges change dynamically and are not fixed; we cannot publish them externally"). Both answers were written to webmasters asking for a whitelist, so the refusal is settled policy rather than an oversight, and the procedure Baidu offers instead is strictly two-step: reverse-lookup the client address, require the hostname to end in `*.baidu.com` or `*.baidu.jp` (「非 *.baidu.com 或 *.baidu.jp 即为冒充」, anything else is an impersonator), then forward-resolve that hostname and require it to come back to the address you started from. What the token actually feeds is Baidu's general web index: the vendor's product table maps it to PC search, mobile search and "other search", with siblings carved off for images, video, news, bookmarks, the ad union and business search. Desktop and mobile have shared the single version identifier `Baiduspider/2.0` since 2015, so the token alone will not tell you which surface is asking: the `android` and `mobile` keywords inside the user agent do.
- Operated by
- Baidu
- Purpose
- Search index
- robots.txt token
- Baiduspider
- Verification
- Vendor-documented
Crawls continuously to build and refresh an AI search index.
How to verify Baiduspider
Baidu publishes no machine-readable range file for Baiduspider that we can fetch on a schedule, so this page shows no range count and no verification date. That is not the same as unverifiable: Baidu documents forward-confirmed reverse DNS, where a genuine request resolves to a hostname under *.baidu.com or *.baidu.jp and that hostname resolves back to the address the request came from. That check is run against one request at a time, at your edge. It is not something we have confirmed on your behalf here.
We fetch no range list for this crawler, so there is no fetch to confirm. This entry was last reviewed against Baidu's own documentation on .
Baidu documents reverse-DNS verification: a genuine request resolves to a hostname under *.baidu.com or *.baidu.jp that forward-resolves back to the same address.
User agent
Baidu publishes this user agent for Baiduspider. Match on the Baiduspider product token rather than the whole string: vendors revise the surrounding version and URL fragments without notice.
Mozilla/5.0 (compatible; Baiduspider/2.0; +http://www.baidu.com/search/spider.html)robots.txt for Baiduspider
Yes, and Baidu restates it emphatically:「百度会严格遵守robots的相关协议」("Baidu will strictly abide by the relevant robots protocols"). It documents `*` and `$` wildcard support inside `Allow` and `Disallow` values and warns that matching is case-exact. Two caveats come from Baidu itself: `Baiduspider-video` "does not support the rules currently", and `Baiduspider-cpro` (ad union) and `Baiduspider-ads` (business search) "will not work on the records set by robots.txt" because they run under customer contracts rather than under robots: those two are opt-out by email, not by directive. Baidu also states that a rule against `Baiduspider` governs all of its user agents.
Block
User-agent: Baiduspider
Disallow: /Allow
User-agent: Baiduspider
Allow: /robots.txt is a request, not an enforcement mechanism. It is honoured by convention, and a crawler that ignores it is stopped at your edge, not in a text file.
What blocking Baiduspider costs you
This is the one entry in its neighbourhood where the trade is conventional organic search rather than AI visibility, and the blast radius is Baidu's own wording: pages "will not be found in Baidu search results and in any other the search results which is provided by Baidu", meaning every syndicated surface rather than baidu.com alone. Because the parent token governs the family, one disallow takes image, video and news search down with it. Reversal is slow and asymmetric: Baidu says a new restriction stops fresh crawling in roughly 48 hours, but already-indexed content takes "several months" to fall out of results, so this is an expensive decision to change your mind about. A site with no Chinese-language audience gives up very little; a site that sells into or recruits from mainland China is closing its primary discovery channel.