diff --git a/containers/pytorch/training/gpu/2.3.0/transformers/4.47.0/py311/Dockerfile b/containers/pytorch/training/gpu/2.3.0/transformers/4.47.0/py311/Dockerfile
index 1ed767f3..bcd5f728 100644
--- a/containers/pytorch/training/gpu/2.3.0/transformers/4.47.0/py311/Dockerfile
+++ b/containers/pytorch/training/gpu/2.3.0/transformers/4.47.0/py311/Dockerfile
@@ -6,7 +6,7 @@ ARG DEBIAN_FRONTEND=noninteractive
 # Versions
 ARG CUDA="cu121"
 ARG PYTORCH="2.3.0"
-ARG FLASH_ATTN="2.7.2"
+ARG FLASH_ATTN="2.6.3"
 ARG TRANSFORMERS="4.47.0"
 ARG HUGGINGFACE_HUB="0.27.0"
 ARG DIFFUSERS="0.31.0"