Skip to content

Free · Multiple files · No signup

Convert CSV to Parquet

Turn large CSV files into compact Parquet files. Choose files to send them for conversion. Follow each file in the list and download each result automatically.

Choose CSV files, get Parquet downloads

1. Choose files → 2. Conversion starts immediately → 3. Download or view your results.

Each file becomes a separate Parquet.

Checking conversion availability and current file limits...

Choosing files starts conversion immediately. Files are sent for processing and automatically deleted after 24 hours. No signup needed.

Three simple steps

How to convert CSV to Parquet

  1. 1. Choose your files

    Add one CSV or a whole batch. The first row supplies the column names.

  2. 2. Convert together

    The list shows each file as it is sent, waits for a turn and converts. Use Add more files while your batch runs.

  3. 3. Save your results

    Each file downloads when ready. Use View for a quick look or Download to save it again.

Ready for analysis

Keep your data, change the format

Create real Apache Parquet files for your data tools. CSV values stay as text, preserving identifiers, leading zeroes and dates as written.

Uploaded CSV files and Parquet results expire 24 hours after conversion starts. You can delete them sooner from the conversion list.

Other ways to convert CSV to Parquet

Prefer a script or command line? These examples run on your computer. They create separate files; review their output rules because external tools can handle types and empty values differently from this page.

Convert CSV to Parquet with Python and PyArrow

Run this on your own computer using Python 3. Save a UTF-8, comma-delimited input.csv in the same folder as the script. The first record must contain unique, nonempty headers; quoted commas and line breaks are parsed as fields.

Before you start

  • Python 3 installed. Package commands below use a local virtual environment on macOS/Linux; on Windows use .venv\Scripts\python.exe in place of .venv/bin/python.
  • Keep the original CSV. These examples read the complete table into memory and create separate output files.
  1. Save the Python code as convert_csv.py beside input.csv.
  2. Install the listed package when required, then run the script. Check the generated output before using it elsewhere.
Install the Python package
python3 -m venv .venv
.venv/bin/python -m pip install pyarrow

Save as convert_csv.py
import csv
from pathlib import Path

with open("input.csv", encoding="utf-8-sig", newline="") as source:
    reader = csv.reader(source, strict=True)
    headers = next(reader)
    rows = list(reader)
if not headers or any(not h for h in headers) or len(set(headers)) != len(headers):
    raise ValueError("Use unique, nonempty column headers")
if any(len(row) != len(headers) for row in rows):
    raise ValueError("Every record must have the same number of fields as the header")

import pyarrow as pa
import pyarrow.parquet as pq
columns = [pa.array([row[i] for row in rows], type=pa.string()) for i in range(len(headers))]
table = pa.Table.from_arrays(columns, names=headers)
pq.write_table(table, "output.parquet", compression="snappy")

Run the conversion
.venv/bin/python convert_csv.py

Limits and output

  • Writes one STRING column per header, preserving leading zeros and empty strings without numeric/date inference. This example rejects blank or duplicate headers; the site uses stable display names. It loads the parsed CSV into memory.

Documentation: Python CSV reader and writer; Apache Arrow Parquet writing

Convert CSV to Parquet with DuckDB

Use DuckDB through Python to read local CSV data and write a separate output file. No database server or upload is required.

Before you start

  • Python 3 and the DuckDB Python package. Run beside input.csv with a header record.
Install DuckDB
python3 -m venv .venv
.venv/bin/python -m pip install duckdb

Save as convert_duckdb.py
import duckdb
from pathlib import Path
if "__CSV_NO_NULL__" in Path("input.csv").read_text(encoding="utf-8-sig"):
    raise ValueError("Choose a null marker absent from your input")
with duckdb.connect() as database:
    database.execute("CREATE TABLE source AS SELECT * FROM read_csv('input.csv', header=true, all_varchar=true, nullstr='__CSV_NO_NULL__', delim=',')")
    database.execute("COPY source TO 'output.parquet' (FORMAT PARQUET);")

Run DuckDB conversion
.venv/bin/python convert_duckdb.py

Limits and output

  • All columns are read as text and empty cells remain empty strings. The reserved null marker is checked before parsing. DuckDB can normalize duplicate or blank column names; review the resulting schema.

Documentation: DuckDB CSV reading options; DuckDB COPY exports

Convert CSV to Parquet with ClickHouse

clickhouse local runs against files on your computer, without starting or connecting to a ClickHouse server.

Before you start

  • Install the ClickHouse binary using the official installation guide. The command assumes clickhouse is on your PATH and input.csv is in the current directory.
  • Choose a new output filename; INTO OUTFILE refuses to overwrite an existing file.
Convert with clickhouse local
clickhouse local --input_format_csv_use_best_effort_in_schema_inference=0 --schema_inference_make_columns_nullable=0 --query "SELECT * FROM file('input.csv', 'CSVWithNames') INTO OUTFILE 'output.parquet' FORMAT Parquet"

Limits and output

  • These settings disable numeric/date inference and nullable columns, preserving empty fields as strings in this example. ClickHouse still has format-specific null markers such as backslash-N; review null settings and output order for your own data rather than assuming identical output to the browser converter.

Documentation: ClickHouse installation; clickhouse-local documentation; ClickHouse CSV input format

Plain answers

CSV to Parquet questions

Can I convert multiple CSV files?

Yes. Add as many files as you need. Each CSV becomes a separate Parquet file, with its own progress, View and Download buttons.

How large can my CSV files be?

The current per-file limit appears beside the file picker. Choose your files and follow their progress in the list.

Will the download include every row?

Yes. Downloads include the complete converted file. View shows a short preview to help you check the result.