If you don't want to download the whole dataset, then you can stream it directly through bcftools and then subset your variants if they are in a list like:

bcftools view -T list.txt https://storage.googleapis.com/gcp-public-data--gnomad/release/2.1.1/vcf/genomes/gnomad.genomes.r2.1.1.sites.vcf.bgz

Or query the annotations you want in the same way using bcftools query

I created a python package based on SQLite databases, where you can easily query all gnomAD variants for GRCh37/38. https://github.com/KalinNonchev/gnomAD_DB I have precomputed SQLite databases for gnomAD WGS for GRCh37/38 in the description of the package. Please take a look there.