Module engine 

Layer 3 complete compute engines.

Each engine is a self-contained GPU compute pipeline: structured input in, compute passes on a real GPU backend, typed output back.

This module owns concrete backend execution helpers. Domain algorithms live in vyre-libs; artifact admission and persistent execution live in vyre-runtime.

Modules

graph

GPU-resident command graph execution. GPU-resident dispatch graph execution (Innovation I.14).

multi_gpu

Mockable multi-GPU work partitioning. Multi-GPU adapter probing, backend acquisition, and work partitioning.