AI Duplicate Test Case Detector

AI Test Optimization

Semantic comparison across a pasted set of test cases identifies exact duplicates, near-duplicates and subsumed cases, with a suggested consolidated list.

Privacy: Requires a free Mutqan account. Your input is sent to the Mutqan AI service to generate the result; prompts and results are not stored — only your account's usage counts for the daily allowance. How the AI tools work

What is a test case?

A test case is a specified set of preconditions, inputs, actions and expected results written to verify one requirement or behaviour (ISTQB). A test scenario is broader — a user goal such as "checkout with an expired card" — that several test cases cover. Good cases are unambiguous, repeatable by someone else, traceable to a requirement, and include the negative and boundary paths, not only the happy one.

Read the guide: How to Write Test Cases

How to use AI Duplicate Test Case Detector

  1. Paste the test cases of the suite — one per line or in any format.
  2. Pick the output language and press Find duplicates.
  3. Review each duplicate group: its kind (exact, near-duplicate or subsumed), the similarity and the recommendation.
  4. Use the consolidated suite as your new list — it covers every original scenario exactly once — or download the groups (CSV) or the JSON.
Only cases with the same intent are grouped, as exact, near-duplicate or subsumed; similar-looking cases that test different data paths stay in the suite.

Exact

The same case, written more than once

Running both:
Checks the same thing twice
In the consolidated suite:
One case, recording the original IDs it replaces

Near-duplicate

The same intent with a small variation

Running both:
Repeats the shared check
In the consolidated suite:
In the example, TC-1 + TC-3 (92%) merge into one case: Successful login (with and without Remember me)

Subsumed

Fully covered by a broader case in the same suite

Running both:
Adds nothing: the broader case already checks it, with the same data
In the consolidated suite:
The broader case, recording the ID it replaces

Not grouped

Similar steps, different data paths: valid vs invalid input

Running both:
Checks two different paths
In the consolidated suite:
Both kept, unchanged

Each group also gets a similarity from 0 to 100 and a recommendation: merge, keep both, delete one or rewrite.

AI Duplicate Test Case Detector features

  • Semantic comparison finds exact, near-duplicate and subsumed cases
  • Similarity percentage and merge/keep/rewrite recommendation per group
  • Consolidated suite that covers every scenario exactly once
  • JSON export for tooling

AI Duplicate Test Case Detector example

Dedupe

Input:

TC-1 Login valid → dashboard; TC-3 Login valid + remember me → dashboard

Output:

DUP-1 near-duplicate (92%) TC-1 + TC-3 → merge: "Successful login (with and without Remember me)"

Frequently asked questions about AI Duplicate Test Case Detector

Are similar-looking tests always duplicates?

No. Cases that exercise different data paths (valid vs invalid input) are kept even when their steps look alike; only same-intent cases are grouped.

What does "subsumed" mean?

The case is fully covered by a broader case in the same suite, so running both adds nothing: everything it checks, the broader case already checks with the same data.

Will it change my test IDs?

No. Existing IDs are kept and used in the groups; only unlabelled cases get IDs (TC-1, TC-2 …) so they can be referred to. The consolidated list records which original IDs each case replaces.

Technical notes

Each group lists the case IDs involved, the kind — exact, near-duplicate, or subsumed by a broader case — a 0–100 similarity, an explanation and a recommendation (merge, keep both, delete one or rewrite), with a merged title when merging. The consolidated list shows which original IDs each remaining case replaces. Unlabelled cases are numbered TC-1, TC-2 … in order.

References