ml-mlip-nvalchemi - Run GPU-Batched MLIP Calculations
Provides optional experimental GPU-batched static prediction, structure relaxation, and molecular dynamics for supported MLIP models using NValchemi.
Tags
Updated: 2026-10-07Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Run batched static predictions
- Run batched structure relaxation
- Run batched molecular dynamics
- Select NValchemi explicitly
- Report execution backends
- Fall back to sequential execution
Inputs
- Structure collections
- Supported MLIP models
- Calculation parameters
- NValchemi selection flag
- GPU resources
Outputs
- Static prediction results
- Relaxed structures
- Molecular dynamics trajectories
- Backend status
- Convergence status
- Runtime log messages
Requirements
- AtomisticSkills runtime
- Compatible MLIP package
- NValchemi toolkit for batch execution
- GPU-compatible runtime for GPU batching