diff --git a/assets/training/finetune_acft_hf_nlp/environments/acpt/context/requirements.txt b/assets/training/finetune_acft_hf_nlp/environments/acpt/context/requirements.txt
index 8efcc45471..b9102b837d 100644
--- a/assets/training/finetune_acft_hf_nlp/environments/acpt/context/requirements.txt
+++ b/assets/training/finetune_acft_hf_nlp/environments/acpt/context/requirements.txt
@@ -6,7 +6,7 @@ azureml-metrics[text]=={{latest-pypi-version}}
 mltable=={{latest-pypi-version}}
 mpi4py==3.1.5
 sentencepiece==0.1.99
-transformers==4.46.3
+transformers==4.48.0
 huggingface-hub==0.25.2
 datasets==2.17.1
 optimum==1.23.3
@@ -24,4 +24,4 @@ trl==0.8.1
 tiktoken==0.6.0
 scipy==1.14.0
 catalogue==2.0.10
-cryptography==43.0.1
\ No newline at end of file
+cryptography==44.0.1
\ No newline at end of file
diff --git a/assets/training/finetune_acft_hf_nlp/environments/data_import/context/Dockerfile b/assets/training/finetune_acft_hf_nlp/environments/data_import/context/Dockerfile
index 3f81fc7bce..28572a3e62 100644
--- a/assets/training/finetune_acft_hf_nlp/environments/data_import/context/Dockerfile
+++ b/assets/training/finetune_acft_hf_nlp/environments/data_import/context/Dockerfile
@@ -16,3 +16,5 @@ RUN pip install scikit-learn==1.5.1
 
 # The below file is required for baking the code into the environment 
 COPY data_import_run.py /azureml/data_import/run.py
+
+# dummy number to change when needing to force rebuild without changing the definition: 1
diff --git a/assets/training/finetune_acft_image/environments/acft_image_huggingface/context/requirements.txt b/assets/training/finetune_acft_image/environments/acft_image_huggingface/context/requirements.txt
index 97824fed8c..0095f52da0 100644
--- a/assets/training/finetune_acft_image/environments/acft_image_huggingface/context/requirements.txt
+++ b/assets/training/finetune_acft_image/environments/acft_image_huggingface/context/requirements.txt
@@ -6,9 +6,9 @@ azureml-core=={{latest-pypi-version}}
 azure-ai-ml==1.19.0
 requests
 datasets==2.19.0
-transformers==4.39.3
+transformers==4.48.0
 accelerate==0.29.3
-optimum==1.19.0
+optimum==1.23.3
 diffusers==0.27.2
 huggingface-hub==0.25.2
 setuptools>=71.1.0
diff --git a/assets/training/finetune_acft_image/environments/acft_image_mmdetection/context/requirements.txt b/assets/training/finetune_acft_image/environments/acft_image_mmdetection/context/requirements.txt
index c3754b650a..5d7bcaa49a 100644
--- a/assets/training/finetune_acft_image/environments/acft_image_mmdetection/context/requirements.txt
+++ b/assets/training/finetune_acft_image/environments/acft_image_mmdetection/context/requirements.txt
@@ -6,9 +6,9 @@ azureml-core=={{latest-pypi-version}}
 azure-ai-ml==1.19.0
 requests
 datasets==2.15.0
-transformers==4.38.2
+transformers==4.48.0
 accelerate==0.27.2
-optimum==1.17.1
+optimum==1.23.3
 diffusers==0.24.0
 huggingface-hub==0.25.2
 openmim==0.3.9
diff --git a/assets/training/finetune_acft_image/environments/acft_video_mmtracking/context/requirements.txt b/assets/training/finetune_acft_image/environments/acft_video_mmtracking/context/requirements.txt
index bfcae5ec74..9a44f5a1a8 100644
--- a/assets/training/finetune_acft_image/environments/acft_video_mmtracking/context/requirements.txt
+++ b/assets/training/finetune_acft_image/environments/acft_video_mmtracking/context/requirements.txt
@@ -6,9 +6,9 @@ azureml-core=={{latest-pypi-version}}
 azure-ai-ml==1.19.0
 requests>=2.31.0
 datasets==2.15.0
-transformers==4.36.2
+transformers==4.48.0
 accelerate==0.25.0
-optimum==1.16.1
+optimum==1.23.3
 diffusers==0.24.0
 openmim==0.3.9
 mmengine==0.8.4
diff --git a/assets/training/finetune_acft_image/environments/acpt_image_framework_selector/context/Dockerfile b/assets/training/finetune_acft_image/environments/acpt_image_framework_selector/context/Dockerfile
index 40dabf3e86..3d283f40a8 100644
--- a/assets/training/finetune_acft_image/environments/acpt_image_framework_selector/context/Dockerfile
+++ b/assets/training/finetune_acft_image/environments/acpt_image_framework_selector/context/Dockerfile
@@ -14,4 +14,4 @@ ENV AZURE_ML_CLI_PRIVATE_FEATURES_ENABLED=True
 RUN pip install gunicorn==22.0.0
 RUN pip install certifi==2024.07.04
 RUN pip install onnx==1.17.0
-# dummy number to change when needing to force rebuild without changing the definition: 2
\ No newline at end of file
+# dummy number to change when needing to force rebuild without changing the definition: 1
\ No newline at end of file
diff --git a/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/Dockerfile b/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/Dockerfile
index d1e21e156b..d16333d723 100644
--- a/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/Dockerfile
+++ b/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/Dockerfile
@@ -14,7 +14,7 @@ RUN pip install azureml-metrics==0.0.33
 # Vulnerability fix
 RUN pip install datasets==2.15.0
 RUN pip install pyarrow==14.0.1
-RUN pip install cryptography==43.0.1
+RUN pip install cryptography==44.0.1
 RUN pip install gunicorn==22.0.0
 RUN pip install onnxruntime==1.17.3
 RUN pip install onnx==1.17.0
diff --git a/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/requirements.txt b/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/requirements.txt
index 02664470a5..36b07498ea 100644
--- a/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/requirements.txt
+++ b/assets/training/finetune_acft_multimodal/environments/acpt_multimodal/context/requirements.txt
@@ -8,10 +8,10 @@ azureml-acft-contrib-hf-nlp=={{latest-pypi-version}}
 # azureml_acft_multimodal_components=={{latest-pypi-version}}
 # needed by multimodal
 mltable==1.5.0
-transformers==4.36.2
+transformers==4.48.0
 peft==0.4.0
 deepspeed==0.15.2
-optimum==1.16.1
+optimum==1.23.3
 accelerate==0.25.0
 diffusers==0.24.0
 onnxruntime-training==1.18.0