JSONL Formatter

updated 25 April 2026

100% client-side. No upload.

Before you start

You need a .jsonl or .ndjson file, or a snippet of JSON Lines text to paste. Technically, this tool also works on regular JSON files, but it's designed to handle multiple JSON objects separated by newlines.

Keep in mind that "Pretty-printed JSONL" is a bit of a contradiction. The JSONL specification requires every record to live on a single line. When you use this tool to "pretty print," it will expand those records across multiple lines for readability. I built this so you can actually read your data, but remember: you'll need to Minify it back down before feeding it into a database or an AI training pipeline.

There is no hard file size limit, but everything happens in your browser's memory. For files over 50 MB, you might notice some lag while the textareas render. If you're dealing with gigabyte-scale logs, a CLI tool is still your best bet.

How to use it

1. Paste your JSONL text into the Input pane, or drop a file onto the page.
2. Set your preferred Indent (usually 2 or 4 spaces).
3. Click Pretty print to expand the JSON objects into a human-readable format.
4. If you're cleaning up a file for production, click Minify to collapse everything back to one line per record.
5. Check the status bar for any line-specific syntax errors.
6. Click Copy or Download to save your work.

Options explained

Indent

This sets the number of spaces used for nesting. I've capped this at 8 spaces to keep the output from flying off the right side of your screen. 0 indent is essentially a "pretty minify" where you get newlines between keys but no horizontal spacing.

Pretty print vs. Minify

Pretty print uses JSON.stringify to add whitespace and newlines within each object. It's great for debugging or manual editing. Minify removes all internal whitespace, ensuring each object occupies exactly one line of text, which is the standard format for NDJSON.

Example

Input (Minified):

{"id":1,"user":"alice","meta":{"v":1}}
{"id":2,"user":"guest","meta":{"v":1}}

Output (Pretty print, 2-space indent):

{
  "id": 1,
  "user": "alice",
  "meta": {
    "v": 1
  }
}

{
  "id": 2,
  "user": "guest",
  "meta": {
    "v": 1
  }
}

Tips & common pitfalls

• The "Pretty" trap. Most JSONL parsers will crash if you give them the pretty-printed version. Always Minify before uploading to OpenAI (use our OpenAI Fine-Tune JSONL Validator for training sets), BigQuery, or Spark.
• Empty lines are ignored. I've designed the parser to skip blank lines, so you don't get "Invalid JSON" errors for the spacing between records.
• Key order is preserved. Modern browsers generally maintain the insertion order of object keys, so your "id" should stay at the top where it belongs.
• Validation is included. If a single line in your file is broken (missing a comma or a quote), the tool will point you to the exact line number—much like the JSONL / NDJSON Viewer—so you can fix it manually.

Troubleshooting

Related tools

• JSONL ValidatorCheck every line for syntax errors and hidden characters.
• JSONL ViewerA spreadsheet-like view for exploring large datasets.
• JSONL ↔ JSON ArrayConvert between one-record-per-line and a single valid JSON array.
• JSONL to CSVFlatten nested JSON records into a flat table for Excel.

See also: if you need to do something adjacent on this site, try JSONL to CSV to flatten JSONL into a CSV with dotted keys, JSONL to JSON to convert JSONL into a JSON array (or back), or OpenAI Fine Tune Validator to validate an OpenAI fine-tune file against the chat schema.

Frequently asked questions