What's the difference between FasterTransformer and TensorRT

[puyuanOT]

Is FasterTransformer developed based on TensorRT?
Is FasterTransformer more efficient than TensorRT when perfoming inference with Transformer models (e.g., llama)?

And what's the difference between FasterTransformer and Huggingface/betterTransformers?

[ankit-db]

#211 (comment)

[duongkstn]

In my case, BetterTransformer from PyTorch is faster than FasterTransformer from NVIDIA (fp32, max len 512, Roberta large)

[jstumpin]

Given latest development, how does this repo fair against TensorRT-LLM?