Free Code Plagiarism Checker API
Fast group similarity checks across 50+ programming languages
Detect code similarity and potential plagiarism with modern, high‑accuracy analysis. Create a free account to get your API key and integrate in minutes.
Start with the API
Create a free account to receive your API key.
A modern MOSS alternative built for accuracy and scale.
Introduction
Since 1997, MOSS (by Stanford) has been a staple for detecting plagiarism in programming courses. Today, with the explosion of open‑source code and AI code generators, copying and remixing code extends far beyond classrooms.
We reimagined the concept with modern algorithms tailored for today’s code ecosystems. Codequiry’s MOSS‑style alternative delivers powerful, accurate group similarity detection with key improvements:
- Multi‑core, horizontally scalable processing for very large datasets.
- Support for 50+ programming languages.
- Advanced similarity analysis that detects structural and semantic edits (renames, reformatting, comment manipulation).
- Optional shared archive to broaden match coverage. Free‑tier submissions may contribute to a collective index to improve detection (you can opt out). Paid plans do not store private code. Our crawlers continuously expand coverage across common code‑sharing and cheating sites.
- Hosted UI with clear visualizations and easy‑to‑interpret JSON results.
- Simple REST API with persistent results available for upgraded keys.
Additionally, our machine‑learning layer identifies AI‑generated code, while our search‑index layer checks billions of code files in real time for potential matches.
Codequiry maintains an extensive archive of web code, enabling unmatched web‑scale plagiarism checks across millions of sites. For source‑code plagiarism detection, our platform is purpose‑built and continuously improving.
Free API: Use our group similarity scanner at no cost. For web and database checks, upgrade to the standard API to unlock full platform access and management tools.
Our infrastructure, backed by high‑end data centers, delivers consistent performance and scale so you can build without licensing constraints or restrictive quotas.
Upload up to 500 ZIP submissions (50+ languages), each up to 10MB, and compute group similarity across the entire set. Note: This product is for group comparisons only and does not perform web checks. The premium platform checks against billions of public sources.
For global plagiarism detection—including the public web, GitHub, Stack Overflow, and billions of additional sources—create an account and explore our full platform. Create a free account to get started.
Our group similarity algorithms are trusted by universities and companies worldwide. They excel at surfacing overlap in large volumes of code—whether analyzing a class of 500 submissions or comparing gigabytes of internal code. Results arrive as clean, structured JSON that’s easy to integrate.
Coming soon: Access our proprietary index of 10+ billion code sources to expand match coverage dramatically.
Want visuals? Create an account to view full web results and interactive dashboards. The free API returns JSON and supports group similarity only. Premium adds real‑time web results across billions of sources.
Limits: We encourage active use of the free API. Free users have a 5‑minute delay between checks (removed when upgrading). Our data centers handle high‑volume workloads; abuse‑based throttling is rare. Free results are deleted after 7 days; upgraded keys unlock persistent storage.
Powering the platform: 20 AMD EPYC™ 9655P servers with a combined 2 TB RAM process thousands of checks in parallel for consistent, low‑latency performance. The Free Group Similarity API supports 50+ languages today and accepts ZIP submissions (each ZIP represents one submission; upload up to 500 per check). The detector parses code natively and remains effective even when code is reformatted, renamed, or commented differently. Web‑scale matching against a multi‑billion‑file index is available on premium, with broader free‑tier coverage coming soon.
All about the account
Before checking code for similarity, create an account using the link below.
Creating an account
Enter billing details and basic information (name, email, role, password). We never sell your data. This information establishes account ownership and lets us provide personalized support.
You can change your account information anytime under "Account Settings" in the dashboard.
Start exploring
After signup and trial activation, explore the dashboard layout. When ready, follow the steps below to run your first check.
Getting started with Codequiry
These tabs contain the most important information on getting started. If you wish to go ahead, the rest is optional!
Folders and creating your first folder
From the dashboard home page, create your first folder to organize checks. Click "New Folder" (top right) and name it (minimum 3 characters).
Checks and creating your first check
From the dashboard, create a check inside a folder. Choose the language parser and select the type of check you wish to run.
Uploading submissions
This step is critical; incorrect uploads can cause incomplete or inaccurate results.
Before uploading, some important information about your files
Provide one .zip per submission containing only that submission’s source code. If you have 10 submissions, upload 10 separate .zip files. A .zip with multiple submissions will be treated as a single submission and will not compare correctly. Ensure each .zip includes at least one file for the selected language (e.g., at least one .java file for Java). We recommend WinZip or WinRAR for compression.
It's time to start checking submissions for plagiarism
When your uploads are ready, click the "Start checking files" button in the Upload/Submissions tab of your check.
Checking types
Great—you’re ready to run checks. Below are the available check types.
After selecting your tests, start checking. Timing depends on file counts, code size, and queue load, but results typically appear instantly or within minutes.
Sharing your account
Share your plan with teammates who need to run checks. Manage access from the "Account Sharing" page in your account settings.
Managing your data
You remain in control of your data. Export and download account‑related data, and view or clear items you no longer wish to retain.
Codequiry does not use your files for checking other users’ submissions and does not share them with third parties.
Interpreting the results
After your check completes, use the "Overview" and "Results" pages to interpret findings and visualizations. Here’s what you’ll see:
-
Match Composition
Shows where matches originate (e.g., GitHub) and the relative share of each source.
-
Overall Variance
Tracks variance between web and group similarity—useful for understanding disparity across larger cohorts.
-
Cluster Graph
Visualizes proximity between submissions. Tight clusters can indicate collaboration or code sharing.
-
Score Bar Chart
Displays each submission with its highest comparison score.
-
Similarity Table
Lists all comparisons and similarity scores. Sort by highest similarity to triage potential issues.
-
File Viewer
Browse parsed source files with inline highlighting of matched segments.
-
Match Explorer
View all matches for the selected file. Peer matches appear first with similarity scores, followed by web matches. Click any match to open side‑by‑side; web matches include the source URL.
-
Plagiarism Score
An estimated probability score based on unoriginality signals—broken down by peer score, web score, and overall average. Treat as guidance and always review underlying matches.
-
Source Makeup
Pie chart estimating composition for the selected submission: unique content, peer‑similar content, and external‑similar content. Use as a directional indicator.
Insights Page
The Insights page organizes estimated similarity across peer and web samples, ranking the highest matches first.
Overview Page
Get a bird’s‑eye view of group similarity scores (Peer Check).
Results Page
Drill into a single submission. See highlighted matches from Peer, Database, and Web checks.
All results are estimates based on similarity thresholds. Codequiry does not determine plagiarism; reviewers should examine matches and context before drawing conclusions.
