Cohere · Training
cohere-ai
Identify cohere-ai by its user agent. Check request identity, review robots.txt controls and track visits in Aiola Analytics.
Key facts
cohere-aiWhat cohere-ai is
Cohere documents no active crawler for foundation-model training and names Coherebot only as a possible future opt-out token. The retained cohere-ai identity is undocumented as a current user agent or robots.txt token.
In Aiola's taxonomy, Training means scraped for model training or dataset development.
cohere-ai user agent
Look for the match token inside the complete HTTP User-Agent header. Tokens can be spoofed, so use the network checks below for authentication.
cohere-aicohere-aiOfficial IP ranges
Cohere has not published official IP ranges that Aiola successfully fetched for cohere-ai. Aiola does not list cloud-provider ranges or community guesses as if they authenticated this crawler.
Verify a cohere-ai request
No IP authentication method is documented in this record for cohere-ai. Its user-agent token is a claim, not proof of identity.
Use the exact header token for classification and apply your own access controls. Check the vendor documentation for any newly published IP, token or header verification method.
Control cohere-ai with robots.txt
Cohere documents crawler controls for this bot family. Robots.txt can direct cooperative crawling, but it cannot prove that a request is authentic.
User-agent: cohere-ai
Allow: /User-agent: cohere-ai
Disallow: /Aiola Analytics
See every cohere-ai visit on your site
Aiola tracks 70 AI, search and link-preview crawlers next to your human traffic. See when cohere-ai arrives, which pages it requests and how that changes over time. Hosted Analytics needs Pro or a trial.
Common questions
Build, ship and fix in one app
Errors, user feedback, GitHub and analytics live next to your coding agents in one desktop app. Every issue becomes a task an agent can take.