[None][feat] modeling_v2: build target C++/CUDA as per-family libraries - #20896

[None][feat] modeling_v2: build target C++/CUDA as per-family libraries#20896

Wanli-Jiang merged 20 commits into

NVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom

Wanli-Jiang:wanli/modeling-v2-native-infraWanli-Jiang/TensorRT-LLM:wanli/modeling-v2-native-infraCopy head branch name to clipboard