Skip to content

Quickstart — Your First CFD Simulation

📦 v0.1.0 — CLI-first. Three products ship: ahmed, drivaer-notchback, onera-m6. These are the only products the CLI, agents, and solvers can interact with today. Product scaffolding and custom geometry are coming in v0.2.0.

Get from zero to a full aerodynamic assessment in under 15 minutes.

1. Install

pip install arxyne

Requires Python 3.12+, Docker, and optionally an NVIDIA GPU (for XLB solver).

2. Doctor

Check your environment:

arxyne --doctor

Verifies: Python, Docker, GPU, solver Docker images, workspace.

Solver images are pulled automatically on first use. No manual Docker setup needed.

3. Pick a Product

arxyne --onboarding

Guided setup — pick a reference product (ahmed, drivaer-notchback, onera-m6). These three products are pre-configured with geometry, solver templates, and validated design-intent targets.

Product scaffolding (--new-product) is gated behind ARXYNE_EXPERIMENTAL=1 and coming properly in v0.2.0. For now, use the bundled reference products:

4. Simulate

arxyne --product ahmed --aero --solver openfoam

Available solvers: openfoam, xlb (GPU LBM), domino (AI surrogate).

Freestream velocity is set per-product in the design_intent layer. Edit products/<name>/layers/design_intent.usda and change arxyne:product:freestream_velocity_mps to your desired speed.

What happens:

  1. Case prep — solver builds mesh/domain from the product's geometry
  2. Simulate — solver runs in Docker (CPU for OpenFOAM, GPU for XLB)
  3. Gates — results evaluated against design-intent targets (Cd, Cl, CmPitch)
  4. Save — results written to the USD digital twin

5. See Results

arxyne --product ahmed --results

Shows current prototype KPIs vs design intent, gate status, and run history. Every simulation builds the traceable audit trail.

6. Run the Full Workflow

arxyne --product ahmed --run-workflow

Auto-dispatches across all domains (aero + dynamics if configured), evaluates KPI gates, auto-promotes baseline when all gates pass.

7. Explore

arxyne --help              # all commands
arxyne --list              # list your products
arxyne --product X --info  # detailed product info

Requirements

Component Required Notes
Python 3.12+ ✅
Docker 24+ ✅ For solver containers
NVIDIA GPU Optional For XLB solver only
NGC account Optional For DoMINO NIM only