Merge branch 'main' into sd-logging-nanobind

.github/workflows/build_packages.yml

@@ -12,6 +12,9 @@ on:
     # Runs at 11:00 AM UTC, which is 3:00 AM PST (UTC-8)
     - cron: '0 11 * * *'
 
+permissions:
+  contents: read
+
 jobs:
   # Note: metadata generation could happen in a separate trigger/schedule
   # workflow. For cross platform builds, it's useful to just generate the
@@ -40,7 +43,7 @@ jobs:
           sharktank_package_version=$(python3 build_tools/python_deploy/compute_local_version.py sharktank)
           shortfin_package_version=$(python3 build_tools/python_deploy/compute_local_version.py shortfin)
       - name: Upload version_local.json
-        uses: actions/upload-artifact@50769540e7f4bd5e21e526ee35c689e35e0d6874 # v4.4.0
+        uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882 # v4.4.3
         with:
           name: version_local
           path: |
@@ -50,6 +53,8 @@ jobs:
   build_packages:
     name: "${{ matrix.package }} :: ${{ matrix.platform }} :: ${{ matrix.python-version }}"
     runs-on: ${{ matrix.runs-on }}
+    permissions:
+      contents: write
     needs: [setup_metadata]
     strategy:
       fail-fast: false
@@ -116,7 +121,7 @@ jobs:
         ./c/shortfin/build_tools/build_linux_package.sh
 
     - name: Upload python wheels
-      uses: actions/upload-artifact@50769540e7f4bd5e21e526ee35c689e35e0d6874 # v4.4.0
+      uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882 # v4.4.3
       with:
         if-no-files-found: error
         name: snapshot-${{ matrix.package }}-${{ matrix.platform }}-${{ matrix.python-version }}
@@ -126,7 +131,6 @@ jobs:
       uses: ncipollo/release-action@2c591bcc8ecdcd2db72b97d6147f871fcd833ba5 # v1.14.0
       with:
         artifacts: bindist/*.whl
-        token: "${{ secrets.RELEASE_PUBLISH_ACCESS_TOKEN }}"
         tag: "dev-wheels"
         name: "dev-wheels"
         body: "Automatic snapshot release of shark-ai python wheels."

.github/workflows/ci-llama-quick-tests.yaml

@@ -78,7 +78,7 @@ jobs:
         run: pytest sharktank/tests/models/llama/benchmark_amdgpu_test.py -v -s --iree-hip-target=gfx942 --run-8b-llama
 
       - name: Upload llama executable files
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@b4b15b8c7c6ac21ea08fcf65892d2ee8f75cf882 # v4.4.3
         with:
           name: llama-files
           path: ${{ github.workspace }}/${{ steps.date.outputs.date }}

docs/user_guide.md

@@ -17,7 +17,7 @@ Our current user guide requires that you have:
 
 This section will help you install Python and set up a Python environment with venv.
 
-Officially we support Python versions: 3.11, 3.12, 3.13, 3.13t
+Officially we support Python versions: 3.11, 3.12, 3.13
 
 The rest of this guide assumes you are using Python 3.11.
 
@@ -39,6 +39,10 @@ Setup your Python environment with the following commands:
 # Set up a virtual environment to isolate packages from other envs.
 python3.11 -m venv 3.11.venv
 source 3.11.venv/bin/activate
+
+# Optional: faster installation of torch with just CPU support.
+# See other options at https://pytorch.org/get-started/locally/
+pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu
 ```
 
 ## Install SHARK and its dependencies