diff --git a/.github/workflows/doctest_job.yml b/.github/workflows/doctest_job.yml
index 86368b0f4c..59b4ce01d5 100644
--- a/.github/workflows/doctest_job.yml
+++ b/.github/workflows/doctest_job.yml
@@ -31,7 +31,7 @@ jobs:
       group: aws-g5-4xlarge-cache
     container:
       image: huggingface/transformers-all-latest-gpu
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     steps:
       - name: Update clone
         working-directory: /transformers
diff --git a/.github/workflows/doctests.yml b/.github/workflows/doctests.yml
index 925701a4c1..9b4bca0ae5 100644
--- a/.github/workflows/doctests.yml
+++ b/.github/workflows/doctests.yml
@@ -18,7 +18,7 @@ jobs:
       group: aws-g5-4xlarge-cache
     container:
       image: huggingface/transformers-all-latest-gpu
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     outputs:
       job_splits: ${{ steps.set-matrix.outputs.job_splits }}
       split_keys: ${{ steps.set-matrix.outputs.split_keys }}
diff --git a/.github/workflows/self-push.yml b/.github/workflows/self-push.yml
index 3ce1ae5599..67aedfc6c0 100644
--- a/.github/workflows/self-push.yml
+++ b/.github/workflows/self-push.yml
@@ -36,7 +36,7 @@ jobs:
       group: '${{ matrix.machine_type }}'
     container:
       image: huggingface/transformers-all-latest-gpu-push-ci
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     outputs:
       matrix: ${{ steps.set-matrix.outputs.matrix }}
       test_map: ${{ steps.set-matrix.outputs.test_map }}
@@ -136,7 +136,7 @@ jobs:
       group: '${{ matrix.machine_type }}'
     container:
       image: huggingface/transformers-all-latest-gpu-push-ci
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     env:
       # For the meaning of these environment variables, see the job `Setup`
       CI_BRANCH_PUSH: ${{ github.event.ref }}
@@ -362,7 +362,7 @@ jobs:
       group: '${{ matrix.machine_type }}'
     container:
       image: huggingface/transformers-pytorch-deepspeed-latest-gpu-push-ci
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     env:
       # For the meaning of these environment variables, see the job `Setup`
       CI_BRANCH_PUSH: ${{ github.event.ref }}
diff --git a/.github/workflows/self-scheduled.yml b/.github/workflows/self-scheduled.yml
index 29b5e3aa14..4b482c28fb 100644
--- a/.github/workflows/self-scheduled.yml
+++ b/.github/workflows/self-scheduled.yml
@@ -55,7 +55,7 @@ jobs:
       group: '${{ matrix.machine_type }}'
     container:
       image: huggingface/transformers-all-latest-gpu
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     outputs:
       folder_slices: ${{ steps.set-matrix.outputs.folder_slices }}
       slice_ids: ${{ steps.set-matrix.outputs.slice_ids }}
@@ -219,7 +219,7 @@ jobs:
       group: '${{ matrix.machine_type }}'
     container:
       image: huggingface/transformers-all-latest-gpu
-      options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+      options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
     steps:
       - name: Update clone
         working-directory: /transformers