Parquet Viewer — Open Parquet Files Online, Privately
View Apache Parquet files in your browser: schema, column types, row count and paged rows. Export Parquet to CSV or JSON without uploading anything.
🔒 Runs entirely in your browser — nothing is uploadedA quick way to look inside Parquet files
Apache Parquet is the default storage format for data lakes, Spark jobs, pandas exports and many analytics pipelines. It is compact and fast, but it is a binary columnar format, so you cannot open it in a text editor or a spreadsheet. This Parquet Viewer opens a .parquet file directly in your browser and shows what is inside: the column names and types, the total number of rows, the number of row groups, the library that wrote the file and a paged table of the actual data.
Under the hood the tool uses DuckDB compiled to WebAssembly. DuckDB reads Parquet natively, including
Snappy, Gzip and Zstd compression, nested lists and structs, dates, timestamps and decimals. The row
count comes from the file metadata, and each page is fetched with a small LIMIT/OFFSET
query, so browsing stays responsive even when the file has millions of rows.
Convert Parquet to CSV or JSON
Often you just need the data in a friendlier format. The export buttons convert the entire file, not only the visible page. CSV export writes a header row followed by every record, ready for Excel, Google Sheets or a database import. JSON export writes a single array of objects, which is convenient for scripts, test fixtures and APIs. Nested columns are kept as nested JSON, while in CSV they appear as text. Timestamps are written in ISO format so they sort and parse correctly.
Private, offline-friendly inspection
Parquet files frequently contain production data: user records, transactions or telemetry. Uploading them to an unknown website just to peek at the schema is risky. Here, the file is read into your browser memory and never sent anywhere. The DuckDB engine (about 35 MB) is downloaded from a CDN the first time you open a file and can then be cached by your browser. Because the whole file is held in memory, very large files depend on the RAM available to your browser tab. For deeper analysis, such as filters, aggregations or joins with other files, open the same file in the SQL on CSV tool.
How to use
- Open a Parquet fileDrop a .parquet file onto the drop area or click to choose one from your computer.
- Inspect the schemaReview column names, DuckDB types, nullability, row count and file metadata.
- Page through the rowsChoose a page size and move between pages or jump to a page number.
- ExportDownload all rows as CSV for spreadsheets or as a JSON array for code.