Skip to content
Delta Grounds
Enter the arena
Public benchmark directory

From drawing comprehension to designs that work.

Architecture, parametric CAD and structural modeling demand different tests. Explore what each benchmark measures, how it grades, and where the evidence stops.

Explore our CAD exam →All enterprise benchmarks
01

Read the drawing

Find dimensions, references and conflicts across documents.

02

Edit the geometry

Make the requested change while preserving protected features.

03

Check the behavior

Test constraints, assemblies or a declared physical response.

  • AEC-Bench196 tasks
    Architecture & CAD

    Drawing navigation and cross-document project coordination. Curated answers and structured checks; some defects are injected into public project records.

    architecturedocument-coordinationstructured checks
    Source reference · integration not included
  • AECV-Bench
    Architecture & CAD

    Drawing perception, OCR, counting and spatial questions. A drawing-reading benchmark rather than a CAD editing environment.

    architecturedrawing-understandingquestion-answer scoring
    Source reference · integration not included
  • DrafterBench1,920 tasks
    Architecture & CAD

    Civil-engineering drawing revision tasks. Grades recorded tool arguments; final CAD geometry is not independently checked by that scoring method.

    architecturedrawing-editingtool-call checks
    Source reference · integration not included
  • BenchCAD
    Architecture & CAD

    Industrial CadQuery generation, reasoning and targeted editing. 17,900 execution-verified programs across 106 families; unseen-family transfer remains difficult.

    cadparametric-editingexecution + geometry
    Source reference · integration not included
  • CADBench · MIT DeCoDE18,000 tasks
    Architecture & CAD

    18,000 evaluation samples across five input modalities. Measures reconstruction, executable programs and compactness. Separate from BenchCAD.

    cadreconstructionexecution + geometry
    Source reference · integration not included
  • CADEngBench600 tasks
    Architecture & CAD

    Parametric parts, controlled edits, assemblies and declared physics checks. 600 part tasks; only 164 parts support full L3 FEA. Solver assumptions are benchmark-assigned.

    cadstructural-engineeringconstraints + FEA
    Source reference · integration not included
  • BMEval / AutoBM128 tasks
    Architecture & CAD

    Simulation-executable building models in OpenSeesPy. Grades execution and period checks against empirical references; passing is not engineering certification.

    structural-engineeringexecution + period checks
    Source reference · integration not included
  • CADGenBench
    Architecture & CAD

    CAD generation and editing with public inputs and private leaderboard reference artifacts. Official grading requires its hosted evaluation setup.

    cadparametric-editingprivate reference grading
    Source reference · integration not included

Directory checked 9 October 2026. Counts and results are author-reported. Source links do not mean these benchmarks are integrated into Delta Grounds or reproduced here. A solver result establishes behavior under the declared assumptions.