YAML to JSON Converter

Converting YAML to JSON here takes three steps, and nothing leaves your browser.

1. 1 Paste or open. Paste YAML into the converter, use Open file for a .yaml or .yml file, or drop the file onto the YAML pane.
2. 2 Pick an indent. Choose 2 spaces, 4 spaces, tab, or minified. The JSON updates as you type.
3. 3 Take it with you. Copy the JSON, download it as a .json file, or make a share link.

If a line isn't valid YAML, the converter tells you the line and column, explains the problem in plain English, and keeps your last valid JSON on screen while you fix it.

Real config files use more than plain keys and values. Here's what happens to the parts JSON doesn't have a direct match for.

Anchors and aliases &defaults *defaults

Expanded into full copies, since JSON has no way to reference another value.

Merge keys <<: *defaults

Merged into the map. Keys you set yourself win over merged ones.

Multiple documents ---

Returned as a JSON array with one item per document, plus a notice so it is never a surprise.

Dates released: 2026-09-24

Written as ISO 8601 strings.

yes, no, on, off enabled: yes

Kept as strings, following YAML 1.2. Only true and false become booleans.

Infinity and NaN .inf -.inf .nan

JSON can’t store these, so they become null, and a warning names the key.

Very large integers id: 9007199254740993

Numbers above 2^53 get a warning, because JavaScript can’t hold them exactly.

Comments # like this

Left out, since JSON has no comment syntax. A one-time note tells you.

YAML is built for people editing config by hand. JSON is built for programs passing data around. Most of the differences follow from that.

Converting in this direction drops YAML-only features: comments disappear, and anchors are written out as full copies. The few values JSON can't store exactly, like .inf, are listed in the section above.

Install PyYAML with pip install pyyaml, then read with safe_load and write with json.dumps. Use safe_load, not load: it only builds plain dicts, lists, strings and numbers, so a file you didn't write can't create arbitrary Python objects.

import json
import yaml

with open("config.yaml") as f:
    data = yaml.safe_load(f)

print(json.dumps(data, indent=2, default=str))

Three ways to do it in a terminal, depending on what's already installed. Each one reads config.yaml and prints JSON, so redirect it with > to save a file.

$ yq -o=json config.yaml > config.json

$ python3 -c 'import sys, json, yaml; print(json.dumps(yaml.safe_load(sys.stdin), indent=2, default=str))' < config.yaml

$ npx js-yaml config.yaml > config.json

mikefarah/yq · tested with yq 4.53.6

Tested with Python 3.11.3 and PyYAML 6.0.1

Tested with Node 22.18.0 and js-yaml 4.3.2

yq from mikefarah is a single binary that also handles JSON, XML and TOML. The Node.js command runs the js-yaml package's own command-line tool through npx, so there is nothing to install globally. It writes dates as full timestamps, like 2026-09-24T00:00:00.000Z.

Yes. The converter is JavaScript running in this tab. Your YAML is parsed in memory and is never sent anywhere, and an automated test checks that on every build. Share links keep the YAML after the # in the address, a part browsers never send to a server.

The only thing the site counts is anonymous page visits, through Cloudflare Web Analytics, which uses no cookies. Your theme and indent choice are saved in your own browser. Your YAML never is.