BCFtools/liftover is a comprehensive and accurate tool designed to convert genomic coordinates of variants across different genome assemblies. It addresses the limitations of existing tools, such as Picard/LiftoverVcf and CrossMap/VCF, which often drop a significant number of variants, especially indels and multi-allelic variants. BCFtools/liftover supports improved handling of indels and multi-allelic variants, updating variant annotation fields when reference alleles change across assemblies. The tool is written in C and is available under the MIT open-source license. It has been evaluated against other tools using the 1000 Genomes project variant callsets, demonstrating lower rates of variant drops and higher accuracy in converting indels and SNVs. BCFtools/liftover is particularly useful for converting variant callsets from large cohorts to novel telomere-to-telomere assemblies and for updating summary statistics from genome-wide association studies tied to legacy genome assemblies. The tool's performance is superior in terms of speed and memory consumption, making it a valuable resource for researchers and clinical laboratories.BCFtools/liftover is a comprehensive and accurate tool designed to convert genomic coordinates of variants across different genome assemblies. It addresses the limitations of existing tools, such as Picard/LiftoverVcf and CrossMap/VCF, which often drop a significant number of variants, especially indels and multi-allelic variants. BCFtools/liftover supports improved handling of indels and multi-allelic variants, updating variant annotation fields when reference alleles change across assemblies. The tool is written in C and is available under the MIT open-source license. It has been evaluated against other tools using the 1000 Genomes project variant callsets, demonstrating lower rates of variant drops and higher accuracy in converting indels and SNVs. BCFtools/liftover is particularly useful for converting variant callsets from large cohorts to novel telomere-to-telomere assemblies and for updating summary statistics from genome-wide association studies tied to legacy genome assemblies. The tool's performance is superior in terms of speed and memory consumption, making it a valuable resource for researchers and clinical laboratories.