Convert a Parquet file to JSON for free, entirely in your browser. Drop in a .parquet file and download it as a JSON array in seconds - the file is never uploaded to a server, and no install or account is needed.
Drop your parquet file here
Everything runs in your browser. Your files stay on your computer.
.parquet file onto the box above.JSON is what most APIs, scripting languages and web tools expect. Unlike CSV it keeps nested structure intact, so structs, lists and maps survive the conversion instead of being flattened into text. That makes it the better target when the data has any shape to it, or when you are feeding it into code rather than a spreadsheet.
Most online converters upload your file, convert it on their server, and email or serve you the result. Parquetly does the conversion in your browser with DuckDB compiled to WebAssembly, so the file is never transmitted. There is no upload, no queue, no size limit imposed by a server, and no copy of your data left behind.
No. The conversion runs in your browser using DuckDB compiled to WebAssembly. Your Parquet file and the JSON it produces both stay on your computer.
A single JSON array of objects, with one object per row and the column names as keys. That is the shape most tools and APIs expect, and it can be parsed in one step.
Yes. Structs become nested objects and lists become arrays, so the structure of the original Parquet file survives the conversion. This is the main advantage of JSON over CSV as a target.
Yes. Use the Query tab to write SQL against the table named data, then download the result. This is useful when the full file would produce more JSON than you want.
Integers and floats become JSON numbers, booleans become true or false, and strings become strings. Structs become objects, lists become arrays, and null stays null. Note that INT64 values beyond about 9 quadrillion are written as JSON numbers, which some JSON parsers read as floating point and may round - if you have identifiers that large, convert to CSV instead, or cast the column to a string in the Query tab first.