NVIDIA/FasterTransformer

Is llama2 70b supported? Do you know minimal configuration?

ChristineSeven opened this issue · 1 comments

Is llama2 70b supported? Do you know minimal configuration?

FasterTransformer development has transitioned to TensorRT-LLM.

LLaMa 2 70B is supported in TensorRT-LLM. Please take a try.