<span class="paragraphSection"><div class="boxTitle">Abstract</div>Post-transcriptional modifications to RNA, which comprise the epitranscriptome, play important roles in RNA metabolism, gene regulation, and disease pathogenesis. However, mapping modifications and characterizing their function is often challenged by a lack of consensus on their presence and significance. The availability of reference epitranscriptomes to benchmark data would significantly advance epitranscriptomic studies. Toward this goal, we established a reference epitranscriptome for human immunodeficiency virus 1 (HIV-1), an important human pathogen. We sequenced a model HIV-1 genome from infected T cells using the latest nanopore technology. A sense and novel preliminary antisense HIV-1 epitranscriptome were generated, where N⁶-methyladenosine, 5-methylcytidine, pseudouridine, inosine, and 2′-<span style="font-style: italic;">O</span>-methyl modifications were mapped by multiplexed base calling at nucleotide resolution. Modification miscalling due to sequence and modification context was corrected with synthetic RNA fragments, and m⁶A was validated with an inhibitor. Modifications were consistent under combination antiretroviral therapy treatment, in primary CD4+ T cells, and in HIV-1 virions. In contrast, spliced transcript-dependent modification levels were observed. Sequencing samples from people living with HIV revealed substantial conservation of m⁶A in circulating strains. Our approach offers a benchmark reference to advance HIV-1 epitranscriptomics and provides a roadmap for the creation of reference epitranscriptomes for other viruses or pathogens.</span>