GPTBot vs PerplexityBot

GPTBot and PerplexityBot compared on 10 documented fields, from 17 claims, last verified 2026-09-13.

Field	GPTBot	PerplexityBot	Comparison
cites_sources	not documented	yes	different
honours_crawl_delay	not_documented	not_documented	same
publishes_ip_list	yes	yes	same
purpose_documented	training	search	different
reads_llms_txt	not documented	not_documented	different
respects_robots_txt	yes	yes	same
robots_txt_fetch_marker	yes	not documented	different
separate_search_and_training_tokens	yes	not_documented	different
user_agent_string_full	Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot	Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot)	different
verifiable_by_rdns	not_documented	not_documented	same

Differences

Field	GPTBot statement	PerplexityBot statement
cites_sources	not documented	Perplexity links to the websites that PerplexityBot surfaces in Perplexity search results.
purpose_documented	GPTBot crawls content that may be used to train OpenAI generative AI foundation models.	PerplexityBot surfaces and links websites in Perplexity search results.
reads_llms_txt	not documented	Perplexity does not publish whether PerplexityBot reads llms.txt files.
robots_txt_fetch_marker	OpenAI may add a robots.txt marker to the GPTBot user agent string when GPTBot fetches robots.txt files.	not documented
separate_search_and_training_tokens	OpenAI separates search and training controls into two robots.txt tokens: OAI-SearchBot for search and GPTBot for training.	Perplexity does not document a separate training crawler token alongside PerplexityBot.
user_agent_string_full	GPTBot sends the user agent string Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko); compatible; GPTBot/1.4; +https://openai.com/gptbot.	PerplexityBot sends the user agent string Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; PerplexityBot/1.0; +https://perplexity.ai/perplexitybot).