Merge pull request #541 from janhq/update-dev-from-master-2026-05-31-01-18

Sync master with upstream release b9437

Author: jan-service-account

.github/actions/ccache-clear/action.yml

@@ -0,0 +1,22 @@
+name: "ccache-clear"
+description: "Delete all GitHub Actions caches matching a key prefix"
+inputs:
+  key:
+    description: "Cache key prefix to match and delete"
+    required: true
+
+runs:
+  using: "composite"
+  steps:
+    - name: Clear caches
+      shell: bash
+      run: |
+        CACHES=$(gh cache list --key "ccache-${{ inputs.key }}" --json id,key --jq '.[] | "\(.id) \(.key)"' 2>/dev/null)
+        if [ -z "$CACHES" ]; then
+          echo "No caches found with key prefix: ${{ inputs.key }}"
+          exit 0
+        fi
+        while read -r id key; do
+          echo "Deleting cache: $id ($key)"
+          gh cache delete "$id"
+        done <<< "$CACHES"

.github/workflows/build-cuda-windows.yml

@@ -13,6 +13,7 @@ concurrency:
   queue: max
 
 env:
+  GH_TOKEN: ${{ github.token }}
   GGML_NLOOP: 3
   GGML_N_THREADS: 1
   LLAMA_ARG_LOG_COLORS: 1
@@ -23,6 +24,9 @@ jobs:
   cuda:
     runs-on: windows-2022
 
+    permissions:
+      actions: write
+
     strategy:
       matrix:
         cuda: ['12.4', '13.3']
@@ -36,7 +40,6 @@ jobs:
         uses: ggml-org/ccache-action@v1.2.21
         with:
           key: release-windows-2022-x64-cuda-${{ matrix.cuda }}
-          append-timestamp: false # note: use this only with non-concurrent jobs!
 
       - name: Install Cuda Toolkit
         uses: ./.github/actions/windows-setup-cuda
@@ -67,9 +70,17 @@ jobs:
           cmake --build build --config Release -j %NINJA_JOBS% -t ggml
           cmake --build build --config Release
 
+      - name: ccache-clear
+        uses: ./.github/actions/ccache-clear
+        with:
+          key: release-windows-2022-x64-cuda-${{ matrix.cuda }}
+
   hip:
     runs-on: windows-2022
 
+    permissions:
+      actions: write
+
     env:
       # Make sure this is in sync with build-cache.yml
       HIPSDK_INSTALLER_VERSION: "26.Q1"
@@ -125,7 +136,6 @@ jobs:
           #       to populate the ccache for the release with manual runs of this workflow
           #key: release-windows-2022-x64-hip-${{ env.HIPSDK_INSTALLER_VERSION }}-${{ matrix.name }}
           key: cuda-windows-2022-x64-hip-${{ env.HIPSDK_INSTALLER_VERSION }}-${{ matrix.name }}
-          append-timestamp: false # note: use this only with non-concurrent jobs!
 
       - name: Build
         id: cmake_build
@@ -144,3 +154,9 @@ jobs:
             -DGPU_TARGETS="gfx1100"  `
             -DGGML_RPC=ON
           cmake --build build -j ${env:NUMBER_OF_PROCESSORS}
+
+      - name: ccache-clear
+        uses: ./.github/actions/ccache-clear
+        with:
+          #key: release-windows-2022-x64-hip-${{ env.HIPSDK_INSTALLER_VERSION }}-${{ matrix.name }}
+          key: cuda-windows-2022-x64-hip-${{ env.HIPSDK_INSTALLER_VERSION }}-${{ matrix.name }}