Decide whether test records must be Jira issues or whether vendor-managed objects fit your governance model.
Jira test management comparison
Xray vs Zephyr vs Nexus
Compare three Jira Cloud test management approaches using current vendor documentation and concrete buying criteria.
Same criteria for all threeSources reviewed 29/09/2026Verify edition and installed release
Side-by-side criteria
Compare the operating models
Capabilities and edition boundaries change. Follow the source links and verify the current contract before purchase.
| Capability | Nexus | Xray | Zephyr |
|---|---|---|---|
| Jira-native testing | ✓ Test Cases, Test Sets and Test Executions are Jira issues. Supporting records use Atlassian Forge storage. | ✓ Core Xray artefacts are Jira work items. Its Marketplace disclosure says specified test details, preconditions and attachments are stored outside Atlassian services. | ✓ Zephyr runs inside Jira. Current Zephyr test cases are Zephyr objects in SmartBear-managed storage; legacy Squad used Jira issues. |
| Test automation integration | ✓ Ingests common report formats and packaged Playwright, Cypress, pytest, GitHub Actions and command-line uploads. | ✓ Imports results from common frameworks and connects to CI/CD through APIs and integrations. | ✓ Supports external CI and BDD integration, plus edition-dependent built-in no-code automation. |
| Customer-hosted automation runner | The optional Nexus runner executes configured commands in customer-controlled infrastructure. Concurrent jobs must use isolated output paths before results are uploaded. | Xray connects to external CI and automation frameworks that execute outside Xray, then imports the results. This is an integration model rather than a Nexus-style runner. | Zephyr supports external CI integrations. Standard and Advanced also offer SmartBear no-code execution, so confirm where each selected automation path runs. |
| Test dataset controls | Nexus models versioned datasets, classifications and parameter rows. Confirm project access controls, reservation behaviour and execution binding in the installed release. | Xray supports parameterised tests, datasets at multiple testing levels and shared parameter value lists. Test Case Designer adds broader model-based data generation. | Zephyr documents test case data, parameters, default values and Call to test reuse of parameterised steps. Evaluate privacy classification, reservation and audit requirements separately. |
| AI test generation | Nexus Pro creates editable drafts from a Jira requirement. Analysis is optional and the organisation connects its chosen supported AI provider. | Current Xray editions offer manual and BDD test generation and prioritisation. Script generation and visual model capabilities depend on edition. | Zephyr offers AI-assisted test steps, Rovo test-case generation and edition-dependent no-code automation capabilities. |
| Review of requirement-generated AI tests | In the requirement-to-test workflow, a person edits, accepts or dismisses each draft before Nexus creates a Jira Test Case. AI-assisted MBT uses a separate direct-creation flow. | Xray describes a human-in-the-loop flow in which generated cases can be reviewed, edited or discarded before creation. Confirm current edition limits. | SmartBear describes Zephyr AI output as ready to review and states that a person controls what enters the test repository. The Rovo agent is documented for Standard and Advanced. |
| Requirement traceability | The traceability report links requirements, cases, executions, environments and defects. Separate dataset bindings and release sign-offs preserve adjacent evidence. | Provides Jira-linked requirements, tests, executions, defects, coverage analysis and reporting. | Provides requirement-to-test traceability, cycles, plans, defects, coverage and reporting inside Jira. |
| Release readiness | Nexus Pro combines configured gates, environment evidence, coverage and recorded sign-off. Current Matrix, open-defect and flaky views use Fix Version scope. | Xray uses plans, executions, environments, coverage and reports to assess readiness to deploy. Confirm the approval and sign-off workflow your team requires. | Zephyr test plans aggregate release progress, results and coverage. SmartBear describes Rovo-assisted readiness for Standard and Advanced; confirm sign-off needs. |
| Editions and pricing | Basic and Pro are the planned Marketplace editions. Confirm production availability and obtain a live quote before making a buying decision. | Standard, Advanced and Enterprise capabilities differ. Use Xray and Atlassian Marketplace pricing for the same tier and required features. | Essential, Standard and Advanced differ in automation and AI capability. Use SmartBear and Atlassian Marketplace pricing for the same tier. |
Buyer validation
Let the workflow decide
Use the same project, requirement, test data and release evidence when trialling each product.
Separate external CI result ingestion from built-in no-code automation and from a customer-hosted job runner.
Test parameterisation with a real dataset, including privacy, concurrency, audit and repeatability requirements.
Generate the same requirement with AI and inspect the editing, approval, provider and usage controls.
Run one representative release and compare evidence currency, exceptions, approvals and final sign-off.
Price the required edition at the same Jira user tier, including migration, AI provider and automation costs.
Comparison sources
Vendor capabilities were reviewed on 29/09/2026. Pricing and editions can change.
- Xray product and edition capabilities
- Xray AI generation and human review
- Xray Cloud architecture and automation documentation
- Xray parameterisation and datasets
- Xray Marketplace privacy and security disclosure
- Zephyr products and edition comparison
- SmartBear comparison of Zephyr and Xray architecture
- Zephyr parameters
- Zephyr AI review and release-readiness workflow
- Zephyr Agent for Rovo edition availability
Questions buyers ask
Frequently asked questions
Which is best: Xray, Zephyr or Nexus?
There is no universal winner. The right choice depends on record architecture, automation model, dataset governance, AI controls, reporting, release workflow, migration and total cost.
Are all three Jira-native?
All three provide test management in Jira. Their record architectures differ: Xray and Nexus use Jira work items for core test artefacts, while current Zephyr documents test cases as Zephyr objects managed by SmartBear.
Do all three support automation?
Yes, but the operating models differ. Compare external CI integration, built-in no-code execution, supported formats and whether a customer-hosted runner is required.
How should we validate the comparison?
Use one representative Jira project, requirement, automated suite, dataset and release. Verify the current edition documentation and contract before purchase.
Are all Nexus capabilities available in every installed version?
No. Availability depends on the installed Nexus release and edition. Confirm the capability in your Jira site during the trial before relying on it.
Test the workflow with a real Jira project
Start with one requirement, one cycle and one release decision.