Build System

The native build produces small shared mechanics plus independently owned family DSOs.

Target	Source owner
trtmc_core	Bundle I/O, device tensors, and engine primitives under core/runtime/.
trtmc_runtime	Exact family/backend loader under core/runtime/loader/.
trtmc_backend_trt	Standard TensorRT Engine implementation.
trtmc_backend_rtx	Optional TensorRT-RTX Engine implementation.
trtmc	Native application under apps/cli/.
trtmc_benchmark_worker	Benchmark application under apps/benchmark/.
trtmc_model_<family>	One family's complete native runtime.

Family discovery

The root CMakeLists.txt glob is limited to families/*/runtime/CMakeLists.txt. Each family file declares its own sources, private dependencies, warnings, tests, output name, and install rule. Adding a normal family never changes a central model source list.

The wheel packages core/builder/tensorrt_model_connect, the top-level families package, benchmark Python code, and installed native binaries/DSOs. Optional family dependencies remain in each families/<family>/requirements.txt; package validation does not import every family implementation.

Typical source build

cmake -S . -B build -G Ninja \
  -DCMAKE_BUILD_TYPE=Release \
  -DCMAKE_CUDA_ARCHITECTURES=100-real \
  -DTRTMC_BUILD_TESTS=OFF \
  -DTRTMC_BUILD_EXAMPLES=OFF
cmake --build build --target \
  trtmc trtmc_backend_trt trtmc_model_qwen

Set the CUDA architecture to the selected GPU. Enable TRTMC_BUILD_BACKEND_RTX only with explicit TensorRT-RTX include and library directories. Disable TRTMC_ENABLE_BYOK only when building without the optional TVM-FFI bridge.

See Build from Source for the complete development-container workflow.