This study aims to describe language variation and language characteristics in the content on @raflychaniago's TikTok account using Chaer and Agustina's sociolinguistic study of language variation theory as the main theory. The method used is a descriptive qualitative method. Data and data sources are in the form of speakers' oral speech containing words, phrases, sentences, and discourse in 4 video content on @raflychaniago's TikTok account. The data collection technique is the free listening technique and the advanced technique of note-taking. The data analysis technique uses the Miles and Huberman data analysis model, namely, researchers collect data, reduce data, present data, draw conclusions, and then verify. The results obtained were found three language variations, namely in terms of speakers consisting of 35 idiolect data; dialect in the form of Ngapak dialect; 3 chronolect data; and 3 types of sociolect, in terms of formality in the form of casual varieties, and in terms of means in the form of spoken language varieties. Researchers also found six characteristics of Rafly's language in his TikTok content.