purge HF_HUB_ENABLE_HF_TRANSFER; promote Xet (#12497)

* purge HF_HUB_ENABLE_HF_TRANSFER; promote Xet

* purge HF_HUB_ENABLE_HF_TRANSFER; promote Xet x2

* restrict docker build test to the ones we actually use in CI.

---------
Co-authored-by: YiYi Xu <yixu310@gmail.com>
Co-authored-by: Sayak Paul <spsayakpaul@gmail.com>

.github/workflows/benchmark.yml

@@ -7,7 +7,7 @@ on:
 
 env:
   DIFFUSERS_IS_CI: yes
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   HF_HOME: /mnt/cache
   OMP_NUM_THREADS: 8
   MKL_NUM_THREADS: 8

.github/workflows/build_docker_images.yml

@@ -42,18 +42,39 @@ jobs:
           CHANGED_FILES: ${{ steps.file_changes.outputs.all }}
         run: |
           echo "$CHANGED_FILES"
+          ALLOWED_IMAGES=(
+            diffusers-pytorch-cpu
+            diffusers-pytorch-cuda
+            diffusers-pytorch-xformers-cuda
+            diffusers-pytorch-minimum-cuda
+            diffusers-doc-builder
+          )
+
+          declare -A IMAGES_TO_BUILD=()
+
           for FILE in $CHANGED_FILES; do
             # skip anything that isn't still on disk
-            if [[ ! -f "$FILE" ]]; then
+            if [[ ! -e "$FILE" ]]; then
               echo "Skipping removed file $FILE"
               continue
             fi
-            if [[ "$FILE" == docker/*Dockerfile ]]; then
-              DOCKER_PATH="${FILE%/Dockerfile}"
-              DOCKER_TAG=$(basename "$DOCKER_PATH")
-              echo "Building Docker image for $DOCKER_TAG"
-              docker build -t "$DOCKER_TAG" "$DOCKER_PATH"
+
+            for IMAGE in "${ALLOWED_IMAGES[@]}"; do
+              if [[ "$FILE" == docker/${IMAGE}/* ]]; then
+                IMAGES_TO_BUILD["$IMAGE"]=1
+              fi
+            done
+          done
+
+          if [[ ${#IMAGES_TO_BUILD[@]} -eq 0 ]]; then
+            echo "No relevant Docker changes detected."
+            exit 0
           fi
+
+          for IMAGE in "${!IMAGES_TO_BUILD[@]}"; do
+            DOCKER_PATH="docker/${IMAGE}"
+            echo "Building Docker image for $IMAGE"
+            docker build -t "$IMAGE" "$DOCKER_PATH"
           done
         if: steps.file_changes.outputs.all != ''
 

.github/workflows/nightly_tests.yml

@@ -7,7 +7,7 @@ on:
 
 env:
   DIFFUSERS_IS_CI: yes
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   OMP_NUM_THREADS: 8
   MKL_NUM_THREADS: 8
   PYTEST_TIMEOUT: 600

.github/workflows/pr_modular_tests.yml

@@ -26,7 +26,7 @@ concurrency:
 
 env:
   DIFFUSERS_IS_CI: yes
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   OMP_NUM_THREADS: 4
   MKL_NUM_THREADS: 4
   PYTEST_TIMEOUT: 60

.github/workflows/pr_tests.yml

@@ -22,7 +22,7 @@ concurrency:
 
 env:
   DIFFUSERS_IS_CI: yes
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   OMP_NUM_THREADS: 4
   MKL_NUM_THREADS: 4
   PYTEST_TIMEOUT: 60

.github/workflows/pr_tests_gpu.yml

@@ -24,7 +24,7 @@ env:
   DIFFUSERS_IS_CI: yes
   OMP_NUM_THREADS: 8
   MKL_NUM_THREADS: 8
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   PYTEST_TIMEOUT: 600
   PIPELINE_USAGE_CUTOFF: 1000000000 # set high cutoff so that only always-test pipelines run
 

.github/workflows/push_tests.yml

@@ -14,7 +14,7 @@ env:
   DIFFUSERS_IS_CI: yes
   OMP_NUM_THREADS: 8
   MKL_NUM_THREADS: 8
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   PYTEST_TIMEOUT: 600
   PIPELINE_USAGE_CUTOFF: 50000
 

.github/workflows/push_tests_fast.yml

@@ -18,7 +18,7 @@ env:
   HF_HOME: /mnt/cache
   OMP_NUM_THREADS: 8
   MKL_NUM_THREADS: 8
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   PYTEST_TIMEOUT: 600
   RUN_SLOW: no
 

.github/workflows/push_tests_mps.yml

@@ -8,7 +8,7 @@ env:
   HF_HOME: /mnt/cache
   OMP_NUM_THREADS: 8
   MKL_NUM_THREADS: 8
-  HF_HUB_ENABLE_HF_TRANSFER: 1
+  HF_XET_HIGH_PERFORMANCE: 1
   PYTEST_TIMEOUT: 600
   RUN_SLOW: no
 

docker/diffusers-doc-builder/Dockerfile

@@ -33,7 +33,7 @@ RUN uv pip install --no-cache-dir "git+https://github.com/huggingface/diffusers.
 RUN uv pip install --no-cache-dir \
     accelerate \
     numpy==1.26.4 \
-    hf_transfer \
+    hf_xet \
     setuptools==69.5.1 \
     bitsandbytes \
     torchao \

docker/diffusers-onnxruntime-cpu/Dockerfile

@@ -44,6 +44,6 @@ RUN python3 -m pip install --no-cache-dir --upgrade pip uv==0.1.11 && \
         scipy \
         tensorboard \
         transformers \
-        hf_transfer
+        hf_xet
 
 CMD ["/bin/bash"]
\ No newline at end of file

docker/diffusers-onnxruntime-cuda/Dockerfile

@@ -38,13 +38,12 @@ RUN python3.10 -m pip install --no-cache-dir --upgrade pip uv==0.1.11 && \
         datasets \
         hf-doc-builder \
         huggingface-hub \
-        hf_transfer \
+        hf_xet \
         Jinja2 \
         librosa \
         numpy==1.26.4 \
         scipy \
         tensorboard \
-        transformers \
-        hf_transfer
+        transformers
 
 CMD ["/bin/bash"]
\ No newline at end of file

docker/diffusers-pytorch-cpu/Dockerfile

@@ -31,7 +31,7 @@ RUN uv pip install --no-cache-dir "git+https://github.com/huggingface/diffusers.
 RUN uv pip install --no-cache-dir \
     accelerate \
     numpy==1.26.4 \
-    hf_transfer
+    hf_xet
 
 RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean
 

docker/diffusers-pytorch-cuda/Dockerfile

@@ -44,6 +44,6 @@ RUN uv pip install --no-cache-dir \
     accelerate \
     numpy==1.26.4 \
     pytorch-lightning \
-    hf_transfer
+    hf_xet
 
 CMD ["/bin/bash"]

docker/diffusers-pytorch-minimum-cuda/Dockerfile

@@ -47,6 +47,6 @@ RUN uv pip install --no-cache-dir \
     accelerate \
     numpy==1.26.4 \
     pytorch-lightning \
-    hf_transfer
+    hf_xet
 
 CMD ["/bin/bash"]

docker/diffusers-pytorch-xformers-cuda/Dockerfile

@@ -44,7 +44,7 @@ RUN uv pip install --no-cache-dir \
     accelerate \
     numpy==1.26.4 \
     pytorch-lightning \
-    hf_transfer \
+    hf_xet \
     xformers
 
 CMD ["/bin/bash"]

tests/quantization/torchao/README.md

@@ -29,7 +29,7 @@ The benchmark results for Flux and CogVideoX can be found in [this](https://gith
 The tests, and the expected slices, were obtained from the `aws-g6e-xlarge-plus` GPU test runners. To run the slow tests, use the following command or an equivalent:
 
 ```bash
-HF_HUB_ENABLE_HF_TRANSFER=1 RUN_SLOW=1 pytest -s tests/quantization/torchao/test_torchao.py::SlowTorchAoTests
+HF_XET_HIGH_PERFORMANCE=1 RUN_SLOW=1 pytest -s tests/quantization/torchao/test_torchao.py::SlowTorchAoTests
 ```
 
 `diffusers-cli`: