One JSON object per line, with no wrapping array. That small change is the whole point: a program can read a record at a time instead of parsing gigabytes into memory first, which is why log pipelines, bulk importers and search indexes all speak it.

Every conversion below runs in this page, on your own machine. Nothing is uploaded, there is no queue, and there is no account to make. Pick a target format, or just drop a NDJSON file above and choose afterwards.