mirror of
https://github.com/NVIDIA/TensorRT-LLM.git
synced 2026-01-14 06:27:45 +08:00
support ring attn for bert_attention plugin and dit model Signed-off-by: ChunhuanLin <lch_xdu@163.com> |
||
|---|---|---|
| .. | ||
| bertAttentionPlugin.cpp | ||
| bertAttentionPlugin.h | ||
| CMakeLists.txt | ||