In the present study, we isolated and characterized the complete mitochondrial genome sequence of Brama japonica by polymerase chain reaction (PCR) amplification and primer-walking sequencing. The complete DNA was 17,009 bp in length and contained a typical set of 13 protein-coding genes, 22 transfer RNA genes, 2 ribosomal RNA genes and a long putative control region. The gene organization and nucleotide composition of complete mitogenome were identical to those of other Bramidae fishes. In contrast, the 12S rRNA gene contained a big poly C structure which was larger than those from other Bramidae species. Of 37 genes, twenty-eight were encoded by heavy strand, while nine were encoded by light strand. Among the 13 protein-coding genes, twelve employed ATG as start codon, while only COI utilized GTG as start codon. In the control region, the terminal associated sequence (TAS), the central and conserved sequence block (CSB-E and CSB-D) and a variable domain (CSB-1, CSB-2 and CSB-3) were identified, while the typical central conserved CSB-F could not be detected in B. japonica. The putative OL region can fold into a conserved secondary structure and the conserved motif (5’-GCCGG-3’) was found at the base of the stem in tRNACys. The overall nucleotide composition of this genome was 26.43% for A, 16.71% for G, 31.35% for C, and 25.50% for T, with a high A+T content of 51.93%. From the NJ phylogenetic tree, we can find that B. japonica was together with other five Bramidae species formed a monophyletic group among 24 species. This work provided a set of useful data for studying on population genetic diversity and molecular evolution in Bramidae and related fish species.