JSON test-file generator

Generate a synthetic JSON file of any size up to 20 GB directly on your machine and stream it to disk. Use the file to benchmark Fast JSON Viewer, your own parser, or any JSON tooling. Nothing is uploaded — the bytes go from your browser straight to a file you pick.

Generate

Tip: 1 KB – 20 GB. Generation runs in a background worker and is typically disk-bound — expect a few hundred MB/s on a fast SSD.

Output format

Top-level JSON object. The first key is always a header so you can identify the file and when it was generated:

{
  "generated": { "by": "fastjsonviewer.com", "on": "<ISO-8601 timestamp>" },
  "<word>_0": <value>,
  "<word>_1": <value>,
  ...
}

Values are a mix of strings (English & French words, multi-byte UTF-8), numbers (integers and floats), booleans, null, and nested objects/arrays up to depth 8. This shape compresses similarly to real-world JSON, so parse-throughput measurements are representative.

How it works

The page uses the File System Access API (showSaveFilePicker) to open a writable stream, then hands the file handle to a Web Worker that appends ~1 MB chunks of JSON until the target size is reached. Generating the next chunk overlaps with writing the previous one, and the page stays responsive throughout. There is no in-memory buffer of the full file — the same code path works for 1 KB and 20 GB.

What to do with the file

• Drag it onto fastjsonviewer.com to test viewer performance on your hardware.
• Feed it to node --experimental-vm-modules, simdjson, jq, etc. and compare parse times.
• Use it as input to load-testing scripts that need realistic JSON payloads at scale.