JSONL to JSON Converter

JSON Lines stores one JSON value per line. It is the format of choice for log pipelines, BigQuery and Elasticsearch exports, and machine-learning datasets — including OpenAI fine-tuning files and many Hugging Face datasets. Many tools, though, expect a single JSON array. This converter validates each line and joins them into one array.

1. Paste JSON Lines or drop a .jsonl / .ndjson file.
2. Each line is parsed separately; an invalid line is reported with its line number.
3. Blank lines are skipped automatically.
4. Copy or download the resulting JSON array.

Example

{"ts": "2026-10-31T08:00:01Z", "level": "info", "msg": "server started", "port": 8080}
{"ts": "2026-10-31T08:00:07Z", "level": "warn", "msg": "slow query", "ms": 812}
{"ts": "2026-10-31T08:01:12Z", "level": "error", "msg": "db timeout", "retry": true}

[
  {
    "ts": "2026-10-31T08:00:01Z",
    "level": "info",
    "msg": "server started",
    "port": 8080
  },
  {
    "ts": "2026-10-31T08:00:07Z",
    "level": "warn",
    "msg": "slow query",
    "ms": 812
  },
  {
    "ts": "2026-10-31T08:01:12Z",
    "level": "error",
    "msg": "db timeout",
    "retry": true
  }
]

Every line becomes one element of the array, in order. Lines do not need to share the same keys.

What is JSON Lines?

A JSON Lines file (also called NDJSON, newline-delimited JSON) is UTF-8 text where every line is a complete, valid JSON value — usually an object — and lines are separated by \n. Because each record stands alone, files can be streamed, appended to and split without parsing the whole file, which is why logging systems and data pipelines love it.

Errors are reported per line

In a large dataset, one broken line can be hard to find. The converter parses line by line and stops at the first invalid one with a message such as Invalid JSON on line 1,284, so you can jump straight to it in your editor. Common culprits are unescaped line breaks inside strings, truncated last lines and pretty-printed records that span several lines.

JSON Lines in AI and data work

• OpenAI and other LLM fine-tuning: each line holds one training example, e.g. a messages array.
• Hugging Face datasets are frequently distributed as .jsonl.
• BigQuery exports and imports newline-delimited JSON.
• Elasticsearch / OpenSearch bulk APIs and many log shippers use NDJSON.

Converting to a JSON array makes the data easy to inspect, or to turn into a table with the JSONL to CSV conversion.

Convert JSON Lines to JSON in code

For one-off files the tool above is quickest. To automate the same task in a script or pipeline, these snippets are a good starting point.

Python

import json

with open("data.jsonl", encoding="utf-8") as f:
    rows = [json.loads(line) for line in f if line.strip()]

with open("data.json", "w", encoding="utf-8") as f:
    json.dump(rows, f, indent=2, ensure_ascii=False)

Command line (jq)

jq -s '.' data.jsonl > data.json

Frequently asked questions

Related converters