Public roadmap

QF Solver 0.2.7 is the current stable source release. This roadmap describes product-level follow-up work; it is not a release gate or a promise that an unqualified route is production-ready.

Current release

This is the 0.2.7 active scope. A historical planning snapshot may retain earlier scores or work-package wording, but it does not define the current release.

The release focuses on inspectable formulations, bounded numerical evidence, reproducible solver behavior and recorded large-model PETSc/MPI workflows. TET4, TET10, HEX8 and HEX20 have the strongest solid-element coverage. WEDGE6 static remains experimental, while its modal claim is limited to the declared first-three-mode scope.

Next technical themes

1. Extend comparable verification for selected element and analysis combinations without broadening claims prematurely.
2. Improve the usability and portability of optional PETSc/MPI workflows.
3. Expand external correlation only where meshes, loads, conventions and observables are demonstrably comparable.
4. Investigate deferred mixed-mesh, WEDGE15, PYRAMID5, HEX8R/SRI/B-bar and finite-kinematic J2 work as separate, evidence-led projects.

Explicitly deferred

General nonlinear/contact production use, finite-sliding production support, GPU claims, universal HPC scaling, 5M Gold and deeper 10M scaling analysis are not part of the current public promise. They require new evidence and a separate decision.

Historical plans and qualification records remain in docs/verification/0_2_7/ for provenance. They are not the active product roadmap.