Skip to content

LeetCode Problems Dataset

This directory documents how to obtain the merged LeetCode problem dataset. The data file merged_problems.json is not committed to this repository; download it into this directory when needed.

  • Local data file: merged_problems.json
  • Source repository: https://github.com/neenza/leetcode-problems/tree/master
  • Original file: https://github.com/neenza/leetcode-problems/blob/master/merged_problems.json
  • Direct download URL: https://raw.githubusercontent.com/neenza/leetcode-problems/master/merged_problems.json

Where This JSON Comes From

merged_problems.json comes from the root directory of neenza/leetcode-problems. That upstream repository contains both a problems/ directory and a pre-merged merged_problems.json file.

This project does not re-crawl or regenerate LeetCode data. When data is needed, download the already merged JSON file from the upstream repository for local reading, analysis, or conversion into solution artifacts.

Data Structure

merged_problems.json contains a top-level questions field, which is an array of problem objects. Each problem object usually contains the following fields:

Some fields (like solutions, images, follow_ups) may be missing for certain problems.

  • title: Problem title, for example Container With Most Water.
  • problem_id: Internal LeetCode problem ID, stored as a string.
  • frontend_id: LeetCode frontend display ID, stored as a string.
  • difficulty: Difficulty level, usually Easy, Medium, or Hard.
  • problem_slug: URL-friendly problem name, for example container-with-most-water.
  • topics: Array of topic tags, such as Array and Two Pointers.
  • description: Full problem statement, usually Markdown or text.
  • examples: Array of examples. Each example object usually contains:
  • example_num: Example number.
  • example_text: Input, output, and explanation text.
  • images: Array of image URLs if the example contains images; this field is not used in model prompts.
  • constraints: Array of problem constraints.
  • follow_ups: Array of follow-up questions, usually empty when none exist.
  • hints: Array of solving hints.
  • code_snippets: Object containing starter code for multiple languages, such as python, cpp, and java.
  • solution / solutions: Official solution or editorial content for some problems, usually as an HTML string.

How to Download or Update

To use the dataset or sync the latest upstream version, run this command from the repository root:

curl -L -o dataset/merged_problems.json https://raw.githubusercontent.com/neenza/leetcode-problems/master/merged_problems.json

After updating, it is recommended to verify that the file is valid JSON and optionally record the upstream commit or download date.