lightseq
lightseq copied to clipboard
性能问题
是否有和tensorrt8.0和fasttransformer4.0做过对比?这写都有一些更新,比如int8推理,比如优化了BS。 另外,有没有实际推理优化的数据,特别是针对大模型,几十亿,上百亿甚至更大的做过试验
There is no update yet to compare with the latest tensorrt and fastertransformer, we will update it in two months LightSeq does not currently support large models