What's New in VectorCAST 2026

Command-line utilities now connect to large language models to generate requirements-based unit tests and code-based requirements. Every generated artifact is fully transparent, with full visibility into inputs, outputs, and mappings at each step before anything enters a verification workflow.

What this means for your program: Teams doing requirements-based testing manually can offload the initial requirement mapping and test case generation to the LLM of their choice, then review and approve before committing. Full traceability is maintained throughout. View more on VectorCAST's AI features here.

Project-level and unit-level coverage now display together in the unit coverage viewer. Previously these required separate views, creating unnecessary context switching when evaluating overall test completeness. Both are now available in the same place.

What this means for your program: Coverage reviews are faster. Assess completeness at both the unit level and the project level without navigating between views.

Existing GoogleTest suites now integrate directly into VectorCAST Coded Tests without rewriting. Execution, coverage analysis, and reporting all work as expected within VectorCAST verification workflows. No migration of test logic required.

What this means for your program: Teams with an established GoogleTest suite can bring it into a certified, coverage-instrumented VectorCAST environment without discarding existing test investment.

Native mock support for C functions is now available inside the xUnit framework. Units under test can be isolated without external mocking mechanisms, keeping test code self-contained and reducing the maintenance surface of your test suite.

What this means for your program: C function mocking no longer requires manual stub management or a separate framework. Test code stays cleaner and the isolation boundary is explicit.

Automatic Test Generation now produces reviewable C++ source code via the --vc-coded-tests option. Generated tests can be version-controlled, reviewed in pull requests, and managed with the same tools as the rest of your codebase.

What this means for your program: ATG output can be checked into source control alongside production code, reviewed in pull requests, and managed with the same tools as the rest of your codebase.