Is it supposed to work with other models supported by TensorRT-LLM

[gggrunt]

I realize it's not drop-in to support other models, do you mind sharing more about some examples to support other models such as Mixtral 8x7B or phi-2 models?

[BLSharda]

Updating to TensorRT-LLM to v0.7.1 will support many models (including mixtral).

For reference of TRT-LLM engine build commands - readme.md from examples<model> folder and it also includes run.py to validate engine with sample input.
https://github.com/NVIDIA/TensorRT-LLM/tree/v0.7.1/examples

pip install tensorrt_llm==0.7.1 --extra-index-url https://pypi.nvidia.com --extra-index-url https://download.pytorch.org/whl/cu121