models: dual-slot backend (qcom/usbgpu) with ref-based downloads (#1966)

* models: dual-slot backend (qcom/usbgpu) with ref-based downloads

* models: restore get_active_source and the usbgpu-to-qcom fallback

* models: fix per-slot validation and cap mismatched-source refetches

* ui/models: select models by ref and seed the usbgpu slot on migration

* models: drop defensive attribute guards on capnp bundles

* models: remove vestigial fetcher state and dead fallbacks

* models: resolve the active bundle from the active source slot only

* models: pass the usbgpu kwarg through the modeld test stubs

* models: resolve the displayed model from the active slot in ui_state

* models: correct the validation memo type hint

* models: drop docstrings that restate the function name

---------

Co-authored-by: Jason Wen <haibin.wen3@gmail.com>

.github/workflows/build-default-big-model.yaml

@@ -1,83 +0,0 @@
-name: Build default big model
-
-on:
-  workflow_dispatch:
-
-env:
-  HF_REPO: sunnypilot/sunnypilot_models_v1
-  HF_DEFAULTS_PATH: models/defaults/big
-
-jobs:
-  resolve_name:
-    runs-on: ubuntu-24.04
-    outputs:
-      model_name: ${{ steps.name.outputs.model_name }}
-      onnx_ref: ${{ steps.name.outputs.onnx_ref }}
-    steps:
-      - uses: actions/checkout@v4
-      - id: name
-        run: |
-          NAME=$(PYTHONPATH=${{ github.workspace }} python3 -c "from openpilot.sunnypilot.models.model_name import DEFAULT_BIG_MODEL; print(DEFAULT_BIG_MODEL)")
-          ONNX_REF=$(git log -1 --format='%H' -- openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx)
-          echo "model_name=${NAME}" >> $GITHUB_OUTPUT
-          echo "onnx_ref=$ONNX_REF" >> $GITHUB_OUTPUT
-
-  build_model:
-    needs: resolve_name
-    uses: ./.github/workflows/sunnypilot-build-model.yaml
-    with:
-      upstream_branch: ${{ needs.resolve_name.outputs.onnx_ref }}
-      custom_name: ${{ needs.resolve_name.outputs.model_name }}
-      target_hardware: usbgpu
-    secrets: inherit
-
-  upload_defaults:
-    needs: [ resolve_name, build_model ]
-    runs-on: ubuntu-24.04
-    permissions:
-      id-token: write
-      contents: write
-    steps:
-      - uses: actions/checkout@v4
-        with:
-          submodules: recursive
-      - run: git lfs pull -I "openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx"
-
-      - name: Install huggingface_hub
-        run: pip install --upgrade "huggingface_hub>=0.22.0"
-
-      - name: Download artifact name
-        uses: actions/download-artifact@v4
-        with:
-          name: artifact-name-${{ needs.resolve_name.outputs.model_name }}
-          path: artifact_name
-
-      - name: Read artifact name
-        id: artifact
-        run: |
-          ARTIFACT_NAME=$(cat artifact_name/artifact_name.txt)
-          echo "artifact_name=$ARTIFACT_NAME" >> $GITHUB_OUTPUT
-
-      - name: Download model artifact
-        uses: actions/download-artifact@v4
-        with:
-          name: ${{ steps.artifact.outputs.artifact_name }}
-          path: output
-
-      - name: Upload to HF and update default_models.json
-        env:
-          HF_OIDC_RESOURCE: datasets/${{ env.HF_REPO }}
-          ARTIFACT_NAME: ${{ steps.artifact.outputs.artifact_name }}
-        run: |
-          rm -f output/artifact_name.txt
-          export PYTHONPATH=$(pwd)
-          python3 release/ci/upload_default_model.py \
-            --hf-repo "${{ env.HF_REPO }}" \
-            --hf-defaults-path "${{ env.HF_DEFAULTS_PATH }}" \
-            --artifact-name "$ARTIFACT_NAME" \
-            --model-dir output \
-            --onnx-path "openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx" \
-            --onnx-ref "${{ needs.resolve_name.outputs.onnx_ref }}" \
-            --model-name "${{ needs.resolve_name.outputs.model_name }}" \
-            --tinygrad-ref "$(python3 openpilot/sunnypilot/models/tinygrad_ref.py)" \
-            --run-number "${{ github.run_number }}"

.github/workflows/build-default-models.yaml

@@ -0,0 +1,501 @@
+name: Build default models
+
+on:
+  workflow_dispatch:
+    inputs:
+      target:
+        description: 'Model target to build'
+        required: true
+        type: choice
+        options:
+          - small
+          - big
+          - dm
+  workflow_call:
+    inputs:
+      target:
+        description: 'Model target to build (small, big, or dm)'
+        required: true
+        type: string
+
+concurrency:
+  group: build-default-models-${{ inputs.target }}
+  cancel-in-progress: false
+
+env:
+  HF_REPO: sunnypilot/sunnypilot_models_v1
+
+jobs:
+  resolve:
+    runs-on: ubuntu-24.04
+    outputs:
+      model_name: ${{ steps.resolve.outputs.model_name }}
+      onnx_ref: ${{ steps.resolve.outputs.onnx_ref }}
+      onnx_path: ${{ steps.resolve.outputs.onnx_path }}
+      hf_defaults_path: ${{ steps.resolve.outputs.hf_defaults_path }}
+      tinygrad_ref: ${{ steps.resolve.outputs.tinygrad_ref }}
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - id: resolve
+        run: |
+          export PYTHONPATH=${{ github.workspace }}
+
+          if [ "${{ inputs.target }}" = "big" ]; then
+            NAME=$(python3 -c "from openpilot.sunnypilot.models.model_name import DEFAULT_BIG_MODEL; print(DEFAULT_BIG_MODEL)")
+            ONNX_PATH="openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx"
+            HF_DEFAULTS_PATH="models/defaults/big"
+          elif [ "${{ inputs.target }}" = "dm" ]; then
+            ONNX_PATH="openpilot/selfdrive/modeld/models/dmonitoring_model.onnx"
+            HF_DEFAULTS_PATH="models/defaults/dm"
+            NAME="dmonitoring_model ($(git log -1 --format=%cd --date=format:'%B %d, %Y' -- "$ONNX_PATH"))"
+          else
+            NAME=$(python3 -c "from openpilot.sunnypilot.models.model_name import DEFAULT_MODEL; print(DEFAULT_MODEL)")
+            ONNX_PATH="openpilot/selfdrive/modeld/models/driving_supercombo.onnx"
+            HF_DEFAULTS_PATH="models/defaults/small"
+          fi
+
+          ONNX_REF=$(git log -1 --format='%H' -- "$ONNX_PATH")
+          TINYGRAD_REF=$(python3 openpilot/sunnypilot/models/tinygrad_ref.py)
+          if [ -z "$TINYGRAD_REF" ]; then
+            echo "::error::Failed to resolve tinygrad ref"
+            exit 1
+          fi
+
+          echo "model_name=${NAME}" >> $GITHUB_OUTPUT
+          echo "onnx_ref=${ONNX_REF}" >> $GITHUB_OUTPUT
+          echo "onnx_path=${ONNX_PATH}" >> $GITHUB_OUTPUT
+          echo "hf_defaults_path=${HF_DEFAULTS_PATH}" >> $GITHUB_OUTPUT
+          echo "tinygrad_ref=${TINYGRAD_REF}" >> $GITHUB_OUTPUT
+
+  build_small_model:
+    needs: resolve
+    if: ${{ inputs.target == 'small' }}
+    runs-on: [self-hosted, tici]
+    env:
+      SMALL_ONNX: openpilot/selfdrive/modeld/models/driving_supercombo.onnx
+      SMALL_PKL: openpilot/selfdrive/modeld/models/driving_tinygrad.pkl
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Pull ONNX via LFS
+        run: git lfs pull -I "${{ env.SMALL_ONNX }}"
+
+      - name: Set environment variables
+        run: |
+          source /etc/profile
+          export UV_PROJECT_ENVIRONMENT=${HOME}/venv
+          export UV_PYTHON_PREFERENCE=managed
+          export UV_PYTHON_INSTALL_DIR=${HOME}/uv/python
+          export VIRTUAL_ENV=$UV_PROJECT_ENVIRONMENT
+          uv sync --frozen
+          printenv >> $GITHUB_ENV
+
+      - name: Disable powersave
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          PYTHONPATH=$PYTHONPATH:${{ github.workspace }}/ ${{ github.workspace }}/scripts/manage-powersave.py --disable
+
+      - name: Compile small model with stock compiler
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH="${PYTHONPATH}:${{ github.workspace }}/tinygrad_repo:${{ github.workspace }}"
+
+          MODEL_SIZE=$(python3 -c "from openpilot.common.transformations.model import MEDMODEL_INPUT_SIZE as s; print(f'{s[0]}x{s[1]}')")
+          CAMERA_RES=$(python3 -c "from openpilot.common.transformations.camera import _ar_ox_fisheye as a, _os_fisheye as o; print(f'{a.width}x{a.height} {o.width}x{o.height}')")
+          FRAME_SKIP=$(python3 -c "from openpilot.selfdrive.modeld.constants import ModelConstants as MC; print(MC.MODEL_RUN_FREQ // MC.MODEL_CONTEXT_FREQ)")
+
+          TG_FLAGS="DEV=QCOM IMAGE=1 FLOAT16=1 NOLOCALS=1 JIT_BATCH_SIZE=0 OPENPILOT_HACKS=1"
+
+          env ${TG_FLAGS} python3 \
+            ${{ github.workspace }}/openpilot/selfdrive/modeld/compile_modeld.py \
+            --onnx ${{ github.workspace }}/${{ env.SMALL_ONNX }} \
+            --model-size $MODEL_SIZE \
+            --camera-resolutions $CAMERA_RES \
+            --frame-skip $FRAME_SKIP \
+            --output ${{ github.workspace }}/${{ env.SMALL_PKL }}
+
+      - name: Chunk small pkl
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH=${{ github.workspace }}
+          python3 -c "
+          from openpilot.common.file_chunker import chunk_file, get_chunk_targets
+          import os
+          pkl = '${{ github.workspace }}/${{ env.SMALL_PKL }}'
+          size = os.path.getsize(pkl)
+          targets = get_chunk_targets(pkl, size)
+          chunk_file(pkl, targets)
+          print(f'Chunked into {len(targets)} files')
+          "
+
+      - name: Prepare output
+        env:
+          MODEL_NAME: ${{ needs.resolve.outputs.model_name }}
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH=${{ github.workspace }}
+          MODELS_DIR="${{ github.workspace }}/openpilot/selfdrive/modeld/models"
+          OUTPUT_DIR="${{ github.workspace }}/small_output"
+          PKL_BASE="driving_tinygrad.pkl"
+          mkdir -p "$OUTPUT_DIR"
+
+          cp "$MODELS_DIR/${PKL_BASE}".chunk* "$OUTPUT_DIR/"
+          cp "$MODELS_DIR/${PKL_BASE}.chunkmanifest" "$OUTPUT_DIR/"
+
+          python3 "${{ github.workspace }}/release/ci/model_generator.py" \
+            --model-dir "$MODELS_DIR" \
+            --output-dir "$OUTPUT_DIR" \
+            --custom-name "$MODEL_NAME" \
+            --upstream-branch "${{ needs.resolve.outputs.onnx_ref }}"
+
+          echo "model-${MODEL_NAME}-${{ github.run_number }}" > "$OUTPUT_DIR/artifact_name.txt"
+
+      - name: Upload small model artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: model-${{ needs.resolve.outputs.model_name }}-${{ github.run_number }}
+          path: ${{ github.workspace }}/small_output/
+
+      - name: Upload artifact name file
+        uses: actions/upload-artifact@v4
+        with:
+          name: artifact-name-${{ needs.resolve.outputs.model_name }}
+          path: ${{ github.workspace }}/small_output/artifact_name.txt
+
+      - name: Re-enable powersave
+        if: always()
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          PYTHONPATH=$PYTHONPATH:${{ github.workspace }}/ ${{ github.workspace }}/scripts/manage-powersave.py --enable
+
+  build_big_model:
+    needs: resolve
+    if: ${{ inputs.target == 'big' }}
+    runs-on: [self-hosted, usbgpu]
+    env:
+      BIG_ONNX: openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx
+      BIG_PKL: openpilot/selfdrive/modeld/models/big_driving_tinygrad.pkl
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Pull big ONNX via LFS
+        run: git lfs pull -I "${{ env.BIG_ONNX }}"
+
+      - name: Set environment variables
+        run: |
+          source /etc/profile
+          export UV_PROJECT_ENVIRONMENT=${HOME}/venv
+          export UV_PYTHON_PREFERENCE=managed
+          export UV_PYTHON_INSTALL_DIR=${HOME}/uv/python
+          export VIRTUAL_ENV=$UV_PROJECT_ENVIRONMENT
+          uv sync --frozen
+          printenv >> $GITHUB_ENV
+
+      - name: Disable powersave
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          PYTHONPATH=$PYTHONPATH:${{ github.workspace }}/ ${{ github.workspace }}/scripts/manage-powersave.py --disable
+
+      - name: Wait for chestnut PCIe link
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH="${PYTHONPATH}:${{ github.workspace }}/tinygrad_repo:${{ github.workspace }}"
+          python3 -c "
+          import time
+          from openpilot.system.hardware.chestnut.flash import link_up
+          for i in range(10):
+              if link_up():
+                  print(f'PCIe link up after {i+1} attempt(s)')
+                  break
+              time.sleep(1)
+          else:
+              raise RuntimeError('Chestnut PCIe link not ready after 10 attempts')
+          "
+
+      - name: Compile big model with stock compiler
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH="${PYTHONPATH}:${{ github.workspace }}/tinygrad_repo:${{ github.workspace }}"
+
+          MODEL_SIZE=$(python3 -c "from openpilot.common.transformations.model import MEDMODEL_INPUT_SIZE as s; print(f'{s[0]}x{s[1]}')")
+          CAMERA_RES=$(python3 -c "from openpilot.common.transformations.camera import _ar_ox_fisheye as a, _os_fisheye as o; print(f'{a.width}x{a.height} {o.width}x{o.height}')")
+          FRAME_SKIP=$(python3 -c "from openpilot.selfdrive.modeld.constants import ModelConstants as MC; print(MC.MODEL_RUN_FREQ // MC.MODEL_CONTEXT_FREQ)")
+
+          TG_FLAGS="DEBUG=2 DEV=USB+AMD:LLVM WARP_DEV=QCOM FLOAT16=1 JIT_BATCH_SIZE=0 GMMU=0 TC_OPT=2"
+
+          env ${TG_FLAGS} python3 \
+            ${{ github.workspace }}/openpilot/selfdrive/modeld/compile_modeld.py \
+            --onnx ${{ github.workspace }}/${{ env.BIG_ONNX }} \
+            --model-size $MODEL_SIZE \
+            --camera-resolutions $CAMERA_RES \
+            --frame-skip $FRAME_SKIP \
+            --output ${{ github.workspace }}/${{ env.BIG_PKL }}
+
+      - name: Chunk big pkl
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH=${{ github.workspace }}
+          python3 -c "
+          from openpilot.common.file_chunker import chunk_file, get_chunk_targets
+          import os
+          pkl = '${{ github.workspace }}/${{ env.BIG_PKL }}'
+          size = os.path.getsize(pkl)
+          targets = get_chunk_targets(pkl, size)
+          chunk_file(pkl, targets)
+          print(f'Chunked into {len(targets)} files')
+          "
+
+      - name: Prepare output
+        env:
+          MODEL_NAME: ${{ needs.resolve.outputs.model_name }}
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH=${{ github.workspace }}
+          MODELS_DIR="${{ github.workspace }}/openpilot/selfdrive/modeld/models"
+          OUTPUT_DIR="${{ github.workspace }}/big_output"
+          PKL_BASE="big_driving_tinygrad.pkl"
+          mkdir -p "$OUTPUT_DIR"
+
+          cp "$MODELS_DIR/${PKL_BASE}".chunk* "$OUTPUT_DIR/"
+          cp "$MODELS_DIR/${PKL_BASE}.chunkmanifest" "$OUTPUT_DIR/"
+
+          python3 "${{ github.workspace }}/release/ci/model_generator.py" \
+            --model-dir "$MODELS_DIR" \
+            --output-dir "$OUTPUT_DIR" \
+            --custom-name "$MODEL_NAME" \
+            --upstream-branch "${{ needs.resolve.outputs.onnx_ref }}"
+
+          echo "model-${MODEL_NAME}-${{ github.run_number }}" > "$OUTPUT_DIR/artifact_name.txt"
+
+      - name: Upload big model artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: model-${{ needs.resolve.outputs.model_name }}-${{ github.run_number }}
+          path: ${{ github.workspace }}/big_output/
+
+      - name: Upload artifact name file
+        uses: actions/upload-artifact@v4
+        with:
+          name: artifact-name-${{ needs.resolve.outputs.model_name }}
+          path: ${{ github.workspace }}/big_output/artifact_name.txt
+
+      - name: Re-enable powersave
+        if: always()
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          PYTHONPATH=$PYTHONPATH:${{ github.workspace }}/ ${{ github.workspace }}/scripts/manage-powersave.py --enable
+
+  upload_defaults:
+    needs: [ resolve, build_small_model, build_big_model, build_dm_model ]
+    if: |
+      ${{
+        !cancelled() &&
+        (inputs.target == 'big' && needs.build_big_model.result == 'success' ||
+         inputs.target == 'small' && needs.build_small_model.result == 'success' ||
+         inputs.target == 'dm' && needs.build_dm_model.result == 'success')
+      }}
+    runs-on: ubuntu-24.04
+    permissions:
+      id-token: write
+      contents: write
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Pull ONNX via LFS
+        run: git lfs pull -I "${{ needs.resolve.outputs.onnx_path }}"
+
+      - name: Install huggingface_hub
+        run: pip install --upgrade "huggingface_hub>=0.22.0"
+
+      - name: Download artifact name
+        if: ${{ inputs.target == 'small' || inputs.target == 'big' }}
+        uses: actions/download-artifact@v4
+        with:
+          name: artifact-name-${{ needs.resolve.outputs.model_name }}
+          path: artifact_name
+
+      - name: Read artifact name
+        if: ${{ inputs.target == 'small' || inputs.target == 'big' }}
+        id: artifact
+        run: |
+          ARTIFACT_NAME=$(cat artifact_name/artifact_name.txt)
+          echo "artifact_name=$ARTIFACT_NAME" >> $GITHUB_OUTPUT
+
+      - name: Download model artifact
+        if: ${{ inputs.target == 'small' || inputs.target == 'big' }}
+        uses: actions/download-artifact@v4
+        with:
+          name: ${{ steps.artifact.outputs.artifact_name }}
+          path: output
+
+      - name: Upload model to HF
+        if: ${{ inputs.target == 'small' || inputs.target == 'big' }}
+        env:
+          HF_OIDC_RESOURCE: datasets/${{ env.HF_REPO }}
+          ARTIFACT_NAME: ${{ steps.artifact.outputs.artifact_name }}
+        run: |
+          rm -f output/artifact_name.txt
+          export PYTHONPATH=$(pwd)
+          python3 release/ci/upload_default_model.py \
+            --hf-repo "${{ env.HF_REPO }}" \
+            --hf-defaults-path "${{ needs.resolve.outputs.hf_defaults_path }}" \
+            --artifact-name "$ARTIFACT_NAME" \
+            --model-dir output \
+            --onnx-path "${{ needs.resolve.outputs.onnx_path }}" \
+            --onnx-ref "${{ needs.resolve.outputs.onnx_ref }}" \
+            --model-name "${{ needs.resolve.outputs.model_name }}" \
+            --tinygrad-ref "${{ needs.resolve.outputs.tinygrad_ref }}" \
+            --run-number "${{ github.run_number }}"
+
+      - name: Download DM artifact
+        if: ${{ inputs.target == 'dm' }}
+        uses: actions/download-artifact@v4
+        with:
+          name: dm-model-${{ github.run_number }}
+          path: dm_output
+
+      - name: Generate DM metadata and upload to HF
+        if: ${{ inputs.target == 'dm' }}
+        env:
+          HF_OIDC_RESOURCE: datasets/${{ env.HF_REPO }}
+        run: |
+          export PYTHONPATH=$(pwd)
+          python3 -c "
+          import json, hashlib
+          from pathlib import Path
+          from datetime import datetime, UTC
+
+          dm_dir = Path('dm_output')
+          manifest = list(dm_dir.glob('*.chunkmanifest'))
+          assert manifest, 'No chunkmanifest found'
+          pkl_name = manifest[0].name.removesuffix('.chunkmanifest')
+          num_chunks = int(manifest[0].read_text().strip())
+
+          chunks = []
+          for i in range(num_chunks):
+              chunk = dm_dir / f'{pkl_name}.chunk{i+1:02d}of{num_chunks:02d}'
+              chunks.append({
+                  'file_name': chunk.name,
+                  'sha256': hashlib.sha256(chunk.read_bytes()).hexdigest()
+              })
+
+          digest = hashlib.sha256()
+          for c in chunks:
+              with open(dm_dir / c['file_name'], 'rb') as f:
+                  while block := f.read(1024*1024):
+                      digest.update(block)
+
+          metadata = {
+              'bundles': [{
+                  'short_name': 'DMMODEL',
+                  'display_name': '${{ needs.resolve.outputs.model_name }}',
+                  'ref': '${{ needs.resolve.outputs.onnx_ref }}',
+                  'runner': 'tinygrad',
+                  'build_time': datetime.now(UTC).strftime('%Y-%m-%dT%H:%M:%SZ'),
+                  'models': [{
+                      'type': 'chunked',
+                      'artifact': {
+                          'file_name': pkl_name,
+                          'download_uri': {'url': '', 'sha256': digest.hexdigest()},
+                          'chunks': chunks
+                      }
+                  }]
+              }]
+          }
+          with open(dm_dir / 'metadata.json', 'w') as f:
+              json.dump(metadata, f, indent=2)
+          print('Generated DM metadata.json')
+          "
+
+          python3 release/ci/upload_default_model.py \
+            --hf-repo "${{ env.HF_REPO }}" \
+            --hf-defaults-path "${{ needs.resolve.outputs.hf_defaults_path }}" \
+            --artifact-name "dm-model-${{ github.run_number }}" \
+            --model-dir dm_output \
+            --onnx-path "${{ needs.resolve.outputs.onnx_path }}" \
+            --onnx-ref "${{ needs.resolve.outputs.onnx_ref }}" \
+            --model-name "${{ needs.resolve.outputs.model_name }}" \
+            --tinygrad-ref "${{ needs.resolve.outputs.tinygrad_ref }}" \
+            --run-number "${{ github.run_number }}"
+
+  build_dm_model:
+    needs: resolve
+    if: ${{ inputs.target == 'dm' }}
+    runs-on: [self-hosted, tici]
+    env:
+      DM_ONNX: openpilot/selfdrive/modeld/models/dmonitoring_model.onnx
+      DM_PKL: openpilot/selfdrive/modeld/models/dmonitoring_model_tinygrad.pkl
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Pull DM ONNX via LFS
+        run: git lfs pull -I "${{ env.DM_ONNX }}"
+
+      - name: Set environment variables
+        run: |
+          source /etc/profile
+          export UV_PROJECT_ENVIRONMENT=${HOME}/venv
+          export UV_PYTHON_PREFERENCE=managed
+          export UV_PYTHON_INSTALL_DIR=${HOME}/uv/python
+          export VIRTUAL_ENV=$UV_PROJECT_ENVIRONMENT
+          uv sync --frozen
+          printenv >> $GITHUB_ENV
+
+      - name: Disable powersave
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          PYTHONPATH=$PYTHONPATH:${{ github.workspace }}/ ${{ github.workspace }}/scripts/manage-powersave.py --disable
+
+      - name: Compile DM model
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH="${PYTHONPATH}:${{ github.workspace }}/tinygrad_repo:${{ github.workspace }}"
+
+          TG_FLAGS="DEV=QCOM IMAGE=1 FLOAT16=1 NOLOCALS=1 JIT_BATCH_SIZE=0 OPENPILOT_HACKS=1"
+
+          taskset -c 7 env ${TG_FLAGS} python3 \
+            ${{ github.workspace }}/tinygrad_repo/examples/openpilot/compile3.py \
+            ${{ github.workspace }}/${{ env.DM_ONNX }} \
+            ${{ github.workspace }}/${{ env.DM_PKL }}
+
+      - name: Chunk DM pkl
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          export PYTHONPATH=${{ github.workspace }}
+          python3 -c "
+          from openpilot.common.file_chunker import chunk_file, get_chunk_targets
+          import os
+          pkl = '${{ github.workspace }}/${{ env.DM_PKL }}'
+          size = os.path.getsize(pkl)
+          targets = get_chunk_targets(pkl, size)
+          chunk_file(pkl, targets)
+          print(f'Chunked {pkl} into {len(targets)} chunks')
+          "
+
+      - name: Prepare DM output
+        run: |
+          mkdir -p dm_output
+          cp ${{ github.workspace }}/${{ env.DM_PKL }}.chunk* dm_output/
+          cp ${{ github.workspace }}/${{ env.DM_PKL }}.chunkmanifest dm_output/
+
+      - name: Upload DM artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: dm-model-${{ github.run_number }}
+          path: dm_output/
+
+      - name: Re-enable powersave
+        if: always()
+        run: |
+          source ${UV_PROJECT_ENVIRONMENT}/bin/activate
+          PYTHONPATH=$PYTHONPATH:${{ github.workspace }}/ ${{ github.workspace }}/scripts/manage-powersave.py --enable
+

.github/workflows/download-hf-model-chunks/action.yml

@@ -0,0 +1,66 @@
+name: Download HF model chunks
+description: Resolve and download model chunks from HuggingFace in parallel
+
+inputs:
+  hf_repo:
+    description: HuggingFace dataset repo
+    required: true
+  models:
+    description: 'JSON array of {hf_path, onnx_hash, canonical} objects'
+    required: true
+  dest_dir:
+    description: Destination directory for downloaded chunks
+    required: true
+
+runs:
+  using: composite
+  steps:
+    - name: Download model chunks
+      shell: bash
+      env:
+        HF_REPO: ${{ inputs.hf_repo }}
+        MODELS_JSON: ${{ inputs.models }}
+        DEST_DIR: ${{ inputs.dest_dir }}
+      run: |
+        set -eo pipefail
+        DOWNLOAD_LIST=$(mktemp)
+
+        resolve_chunks() {
+          local HF_PATH="$1" ONNX_HASH="$2" CANONICAL="$3" DEST_DIR="$4"
+          local JSON_URL="https://huggingface.co/datasets/${HF_REPO}/resolve/main/${HF_PATH}/default_models.json"
+          local DEFAULTS BUNDLE ARTIFACT BASE_URL NUM_CHUNKS
+          DEFAULTS=$(curl -fsSL "$JSON_URL")
+          BUNDLE=$(echo "$DEFAULTS" | jq --arg hash "$ONNX_HASH" '.bundles[] | select(.onnx_sha256 == $hash)')
+          ARTIFACT=$(echo "$BUNDLE" | jq -r '.models[0].artifact')
+          BASE_URL=$(echo "$ARTIFACT" | jq -r '.download_uri.url' | sed 's|/[^/]*$||')
+          NUM_CHUNKS=$(echo "$ARTIFACT" | jq -r '.chunks | length')
+
+          mkdir -p "$DEST_DIR"
+          while IFS= read -r CHUNK_NAME; do
+            CHUNK_IDX=$(echo "$CHUNK_NAME" | grep -oP 'chunk\K[0-9]+of[0-9]+' || true)
+            if [ -z "$CHUNK_IDX" ]; then
+              echo "::error::Failed to parse chunk index from: $CHUNK_NAME"
+              return 1
+            fi
+            ENCODED_URL=$(python3 -c "import urllib.parse; print(urllib.parse.quote('${BASE_URL}/${CHUNK_NAME}', safe=':/'))")
+            printf '%s\t%s\n' "$ENCODED_URL" "${DEST_DIR}/${CANONICAL}.chunk${CHUNK_IDX}" >> "$DOWNLOAD_LIST"
+          done < <(echo "$ARTIFACT" | jq -r '.chunks[].file_name')
+          echo "$NUM_CHUNKS" > "${DEST_DIR}/${CANONICAL}.chunkmanifest"
+        }
+
+        echo "$MODELS_JSON" | jq -c '.[]' | while IFS= read -r model; do
+          HF_PATH=$(echo "$model" | jq -r '.hf_path')
+          ONNX_HASH=$(echo "$model" | jq -r '.onnx_hash')
+          CANONICAL=$(echo "$model" | jq -r '.canonical')
+          resolve_chunks "$HF_PATH" "$ONNX_HASH" "$CANONICAL" "$DEST_DIR"
+        done
+
+        TOTAL=$(wc -l < "$DOWNLOAD_LIST")
+        echo "Downloading $TOTAL chunks with 8 parallel connections..."
+        xargs -P8 -d'\n' -I{} bash -c '
+          URL="${1%%	*}"
+          DEST="${1#*	}"
+          echo "Downloading $(basename "$DEST")"
+          curl -fsSL --retry 3 --retry-delay 5 -o "$DEST" "$URL"
+        ' _ {} < "$DOWNLOAD_LIST"
+        rm -f "$DOWNLOAD_LIST"

.github/workflows/sunnypilot-build-model.yaml

@@ -188,7 +188,7 @@ jobs:
           if [ "${{ inputs.target_hardware }}" == "usbgpu" ]; then
             echo "USBGPU build"
             export USBGPU=1
-            TG_FLAGS="DEBUG=2 DEV=USB+AMD:LLVM WARP_DEV=QCOM FLOAT16=1 JIT_BATCH_SIZE=0 GMMU=0 TC_OPT=2"
+            TG_FLAGS="DEBUG=1 DEV=USB+AMD:LLVM WARP_DEV=QCOM FLOAT16=1 JIT_BATCH_SIZE=0 GMMU=0 TC_OPT=2"
             OUTPUT_PKL="${{ env.MODELS_DIR }}/big_driving_tinygrad.pkl"
           else
             echo "QCOM build"

.github/workflows/sunnypilot-build-prebuilt.yaml

@@ -39,6 +39,8 @@ jobs:
       include_big_model: ${{ steps.strategy.outputs.include_big_model }}
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
       - name: Extract deploy strategy
         id: strategy
         run: |
@@ -96,6 +98,8 @@ jobs:
       }}
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
       - name: Wait for Tests
         uses: ./.github/workflows/wait-for-action # Path to where you place the action
         with:
@@ -119,6 +123,7 @@ jobs:
     steps:
       - uses: actions/checkout@v4
         with:
+          fetch-depth: 1
           submodules: recursive
           ref: ${{ env.SOURCE_BRANCH }}
           repository: ${{ github.event.pull_request.head.repo.fork && github.event.pull_request.head.repo.full_name || github.repository }}
@@ -165,7 +170,7 @@ jobs:
           scons -j1 cache_dir="$SCONS_CACHE" --minimal \
             openpilot/selfdrive/locationd openpilot/sunnypilot/selfdrive/locationd
           echo "Building rest of sunnypilot"
-          /usr/bin/time -v scons -j$(nproc) cache_dir="$SCONS_CACHE" --minimal
+          SKIP_TINYGRAD_COMPILE=1 /usr/bin/time -v scons -j$(nproc) cache_dir="$SCONS_CACHE" --minimal
           touch ${BUILD_DIR}/prebuilt
           if [[ "${{ runner.debug }}" == "1" ]]; then
             ls -la ${BUILD_DIR}
@@ -214,59 +219,174 @@ jobs:
     outputs:
       onnx_sha256: ${{ steps.resolve.outputs.onnx_sha256 }}
     env:
+      GH_REPO: ${{ github.repository }}
       HF_REPO: sunnypilot/sunnypilot_models_v1
       HF_DEFAULTS_PATH: models/defaults/big
     steps:
-      - uses: actions/checkout@v4
-        with:
-          ref: ${{ github.head_ref || github.ref_name }}
-
-      - run: git lfs pull -I "openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx"
-
-      - name: Check HF defaults and build if needed
+      - name: Resolve ONNX hash and tinygrad ref via API
         id: resolve
         run: |
-          ACTUAL_ONNX_HASH=$(sha256sum "openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx" | cut -d' ' -f1)
-          echo "Repo ONNX hash: $ACTUAL_ONNX_HASH"
-          echo "onnx_sha256=$ACTUAL_ONNX_HASH" >> $GITHUB_OUTPUT
+          REF="${{ github.head_ref || github.ref_name }}"
+
+          ONNX_HASH=$(gh api "repos/${GH_REPO}/contents/openpilot/selfdrive/modeld/models/big_driving_supercombo.onnx?ref=${REF}" --jq '.content' | base64 -d | grep '^oid sha256:' | cut -d: -f2)
+          echo "ONNX hash: $ONNX_HASH"
+          echo "onnx_sha256=$ONNX_HASH" >> $GITHUB_OUTPUT
+
+          TINYGRAD_REF=$(gh api "repos/${GH_REPO}/contents/tinygrad_repo?ref=${REF}" --jq '.sha')
+          echo "tinygrad ref: $TINYGRAD_REF"
 
           JSON_URL="https://huggingface.co/datasets/${HF_REPO}/resolve/main/${HF_DEFAULTS_PATH}/default_models.json"
 
-          check_hash() {
+          check_defaults() {
             DEFAULTS=$(curl -fsSL "$JSON_URL" 2>/dev/null) || return 1
-            BUNDLE=$(echo "$DEFAULTS" | jq --arg hash "$ACTUAL_ONNX_HASH" '.bundles[] | select(.onnx_sha256 == $hash)' 2>/dev/null)
+            TINYGRAD_MATCH=$(echo "$DEFAULTS" | jq -r --arg ref "$TINYGRAD_REF" '.tinygrad_ref == $ref' 2>/dev/null)
+            [ "$TINYGRAD_MATCH" = "true" ] || return 1
+            BUNDLE=$(echo "$DEFAULTS" | jq --arg hash "$ONNX_HASH" '.bundles[] | select(.onnx_sha256 == $hash)' 2>/dev/null)
             [ -n "$BUNDLE" ] && [ "$BUNDLE" != "null" ]
           }
 
-          if check_hash; then
-            echo "HF defaults match repo ONNX"
-          else
-            echo "No matching model on HF — triggering build"
-            gh workflow run build-default-big-model.yaml --ref "${{ github.head_ref || github.ref_name }}"
-
-            echo "Waiting for build to start..."
-            sleep 120
-
-            RUN_ID=$(gh run list --workflow=build-default-big-model.yaml --branch="${{ github.head_ref || github.ref_name }}" --limit=1 --json databaseId --jq '.[0].databaseId')
-            if [ -z "$RUN_ID" ] || [ "$RUN_ID" = "null" ]; then
-              echo "::error::Failed to find build-default-big-model run"
-              exit 1
-            fi
-
-            echo "Waiting for run $RUN_ID..."
-            gh run watch "$RUN_ID"
-
-            CONCLUSION=$(gh run view "$RUN_ID" --json conclusion --jq '.conclusion')
-            if [ "$CONCLUSION" != "success" ]; then
-              echo "::error::build-default-big-model failed: $CONCLUSION"
-              exit 1
-            fi
-
-            if ! check_hash; then
-              echo "::error::HF defaults still don't match after build"
-              exit 1
-            fi
+          if check_defaults; then
+            echo "HF defaults match repo ONNX hash and tinygrad ref"
+            exit 0
           fi
+
+          echo "No matching model on HF — dispatching build"
+          gh workflow run build-default-models.yaml --ref "$REF" -f target=big
+
+          echo "Polling HF for big model availability..."
+          for i in $(seq 1 90); do
+            sleep 30
+            if check_defaults; then
+              echo "Big model available on HF after $((i * 30))s"
+              exit 0
+            fi
+            echo "Poll $i/90: not yet available"
+          done
+
+          echo "::error::Big model not available on HF after 45 minutes"
+          exit 1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Cancel run on failure
+        if: failure()
+        run: gh run cancel ${{ github.run_id }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+  prepare_small_model:
+    needs: [ prepare_strategy ]
+    runs-on: ubuntu-24.04
+    outputs:
+      driving_onnx_sha256: ${{ steps.resolve.outputs.driving_onnx_sha256 }}
+    env:
+      GH_REPO: ${{ github.repository }}
+      HF_REPO: sunnypilot/sunnypilot_models_v1
+      HF_DEFAULTS_PATH: models/defaults/small
+    steps:
+      - name: Resolve ONNX hash and tinygrad ref via API
+        id: resolve
+        run: |
+          REF="${{ github.head_ref || github.ref_name }}"
+
+          DRIVING_HASH=$(gh api "repos/${GH_REPO}/contents/openpilot/selfdrive/modeld/models/driving_supercombo.onnx?ref=${REF}" --jq '.content' | base64 -d | grep '^oid sha256:' | cut -d: -f2)
+          echo "Driving ONNX hash: $DRIVING_HASH"
+          echo "driving_onnx_sha256=$DRIVING_HASH" >> $GITHUB_OUTPUT
+
+          TINYGRAD_REF=$(gh api "repos/${GH_REPO}/contents/tinygrad_repo?ref=${REF}" --jq '.sha')
+          echo "tinygrad ref: $TINYGRAD_REF"
+
+          JSON_URL="https://huggingface.co/datasets/${HF_REPO}/resolve/main/${HF_DEFAULTS_PATH}/default_models.json"
+
+          check_defaults() {
+            DEFAULTS=$(curl -fsSL "$JSON_URL" 2>/dev/null) || return 1
+            TINYGRAD_MATCH=$(echo "$DEFAULTS" | jq -r --arg ref "$TINYGRAD_REF" '.tinygrad_ref == $ref' 2>/dev/null)
+            [ "$TINYGRAD_MATCH" = "true" ] || return 1
+            DRIVING=$(echo "$DEFAULTS" | jq --arg hash "$DRIVING_HASH" '.bundles[] | select(.onnx_sha256 == $hash)' 2>/dev/null)
+            [ -n "$DRIVING" ] && [ "$DRIVING" != "null" ] || return 1
+          }
+
+          if check_defaults; then
+            echo "HF defaults match repo ONNX hash and tinygrad ref"
+            exit 0
+          fi
+
+          echo "No matching model on HF — dispatching build"
+          gh workflow run build-default-models.yaml --ref "$REF" -f target=small
+
+          echo "Polling HF for model availability..."
+          for i in $(seq 1 60); do
+            sleep 30
+            if check_defaults; then
+              echo "Model available on HF after $((i * 30))s"
+              exit 0
+            fi
+            echo "Poll $i/60: not yet available"
+          done
+
+          echo "::error::Small driving model not available on HF after 30 minutes"
+          exit 1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Cancel run on failure
+        if: failure()
+        run: gh run cancel ${{ github.run_id }}
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+  prepare_dm_model:
+    needs: [ prepare_strategy ]
+    runs-on: ubuntu-24.04
+    outputs:
+      dm_onnx_sha256: ${{ steps.resolve.outputs.dm_onnx_sha256 }}
+    env:
+      GH_REPO: ${{ github.repository }}
+      HF_REPO: sunnypilot/sunnypilot_models_v1
+      HF_DEFAULTS_PATH: models/defaults/dm
+    steps:
+      - name: Resolve ONNX hash and tinygrad ref via API
+        id: resolve
+        run: |
+          REF="${{ github.head_ref || github.ref_name }}"
+
+          DM_HASH=$(gh api "repos/${GH_REPO}/contents/openpilot/selfdrive/modeld/models/dmonitoring_model.onnx?ref=${REF}" --jq '.content' | base64 -d | grep '^oid sha256:' | cut -d: -f2)
+          echo "DM ONNX hash: $DM_HASH"
+          echo "dm_onnx_sha256=$DM_HASH" >> $GITHUB_OUTPUT
+
+          TINYGRAD_REF=$(gh api "repos/${GH_REPO}/contents/tinygrad_repo?ref=${REF}" --jq '.sha')
+          echo "tinygrad ref: $TINYGRAD_REF"
+
+          JSON_URL="https://huggingface.co/datasets/${HF_REPO}/resolve/main/${HF_DEFAULTS_PATH}/default_models.json"
+
+          check_defaults() {
+            DEFAULTS=$(curl -fsSL "$JSON_URL" 2>/dev/null) || return 1
+            TINYGRAD_MATCH=$(echo "$DEFAULTS" | jq -r --arg ref "$TINYGRAD_REF" '.tinygrad_ref == $ref' 2>/dev/null)
+            [ "$TINYGRAD_MATCH" = "true" ] || return 1
+            DM=$(echo "$DEFAULTS" | jq --arg hash "$DM_HASH" '.bundles[] | select(.onnx_sha256 == $hash)' 2>/dev/null)
+            [ -n "$DM" ] && [ "$DM" != "null" ] || return 1
+          }
+
+          if check_defaults; then
+            echo "HF defaults match DM ONNX hash and tinygrad ref"
+            exit 0
+          fi
+
+          echo "No matching DM model on HF — dispatching build"
+          gh workflow run build-default-models.yaml --ref "$REF" -f target=dm
+
+          echo "Polling HF for DM model availability..."
+          for i in $(seq 1 60); do
+            sleep 30
+            if check_defaults; then
+              echo "DM model available on HF after $((i * 30))s"
+              exit 0
+            fi
+            echo "Poll $i/60: not yet available"
+          done
+
+          echo "::error::DM model not available on HF after 30 minutes"
+          exit 1
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
@@ -278,23 +398,24 @@ jobs:
 
   publish:
     concurrency:
-      # We do a bit of a hack here to avoid canceling the publishing job if a new commit comes in while we're publishing by adding the sha to the group name.
-      # This means that if multiple commits come in while we're publishing, they will be queued up and publish one after the other.
-      # Otherwise, if a job is waiting to be published due to environment wait time, it would be canceled by a new commit and restart the wait time.
       group: ${{ needs.prepare_strategy.outputs.publish_concurrency_group }}
       cancel-in-progress: ${{ needs.prepare_strategy.outputs.cancel_publish_in_progress == 'true' }}
     if: ${{
       always() && !cancelled() &&
       needs.build.result == 'success' &&
       needs.prepare_strategy.result == 'success' &&
+      needs.prepare_small_model.result == 'success' &&
+      needs.prepare_dm_model.result == 'success' &&
       (!contains(github.event_name, 'pull_request') || (github.event.action == 'labeled' && github.event.label.name == 'prebuilt')) &&
       (needs.prepare_strategy.outputs.include_big_model != 'true' || needs.prepare_chestnut.result == 'success')
       }}
-    needs: [ build, prepare_strategy, prepare_chestnut ]
+    needs: [ build, prepare_strategy, prepare_chestnut, prepare_small_model, prepare_dm_model ]
     runs-on: ubuntu-24.04
     environment: ${{ needs.prepare_strategy.outputs.environment }}
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
 
       - name: Download prebuilt artifact
         uses: actions/download-artifact@v4
@@ -306,43 +427,16 @@ jobs:
           mkdir -p ${{ env.OUTPUT_DIR }}
           tar xzf prebuilt.tar.gz -C ${{ env.OUTPUT_DIR }}
 
-      - name: Prepare chestnut output
-        if: ${{ needs.prepare_chestnut.result == 'success' }}
-        run: |
-          mkdir -p "${{ github.workspace }}/chestnut_output"
-          tar xzf prebuilt.tar.gz -C "${{ github.workspace }}/chestnut_output"
-
-      - name: Download big model chunks from HF
-        if: ${{ needs.prepare_chestnut.result == 'success' }}
-        env:
-          HF_REPO: sunnypilot/sunnypilot_models_v1
-          HF_DEFAULTS_PATH: models/defaults/big
-        run: |
-          ONNX_HASH="${{ needs.prepare_chestnut.outputs.onnx_sha256 }}"
-          JSON_URL="https://huggingface.co/datasets/${HF_REPO}/resolve/main/${HF_DEFAULTS_PATH}/default_models.json"
-          DEFAULTS=$(curl -fsSL "$JSON_URL")
-          BUNDLE=$(echo "$DEFAULTS" | jq --arg hash "$ONNX_HASH" '.bundles[] | select(.onnx_sha256 == $hash)')
-
-          mkdir -p big_model_chunks
-          ARTIFACT=$(echo "$BUNDLE" | jq -r '.models[0].artifact')
-          BASE_URL=$(echo "$ARTIFACT" | jq -r '.download_uri.url' | sed 's|/[^/]*$||')
-          NUM_CHUNKS=$(echo "$ARTIFACT" | jq -r '.chunks | length')
-
-          CANONICAL="big_driving_tinygrad.pkl"
-          echo "$ARTIFACT" | jq -r '.chunks[].file_name' | while read CHUNK_NAME; do
-            CHUNK_IDX=$(echo "$CHUNK_NAME" | grep -oP 'chunk\K[0-9]+of[0-9]+')
-            CANONICAL_CHUNK="${CANONICAL}.chunk${CHUNK_IDX}"
-            ENCODED_URL=$(python3 -c "import urllib.parse; print(urllib.parse.quote('${BASE_URL}/${CHUNK_NAME}', safe=':/'))")
-            echo "Downloading $CHUNK_NAME -> $CANONICAL_CHUNK"
-            curl -fsSL -o "big_model_chunks/${CANONICAL_CHUNK}" "$ENCODED_URL"
-          done
-
-          echo "$NUM_CHUNKS" > "big_model_chunks/${CANONICAL}.chunkmanifest"
-
-      - name: Inject big model into chestnut
-        if: ${{ needs.prepare_chestnut.result == 'success' }}
-        run: |
-          cp big_model_chunks/* "${{ github.workspace }}/chestnut_output/openpilot/selfdrive/modeld/models/"
+      - name: Download model chunks from HF
+        uses: ./.github/workflows/download-hf-model-chunks
+        with:
+          hf_repo: sunnypilot/sunnypilot_models_v1
+          dest_dir: ${{ env.OUTPUT_DIR }}/openpilot/selfdrive/modeld/models
+          models: |
+            [
+              {"hf_path": "models/defaults/small", "onnx_hash": "${{ needs.prepare_small_model.outputs.driving_onnx_sha256 }}", "canonical": "driving_tinygrad.pkl"},
+              {"hf_path": "models/defaults/dm", "onnx_hash": "${{ needs.prepare_dm_model.outputs.dm_onnx_sha256 }}", "canonical": "dmonitoring_model_tinygrad.pkl"}
+            ]
 
       - name: Configure Git
         run: |
@@ -364,22 +458,6 @@ jobs:
             "https://x-access-token:${{github.token}}@github.com/sunnypilot/sunnypilot.git" \
             "${{ needs.prepare_strategy.outputs.extra_version_identifier }}"
 
-      - name: Publish chestnut branch
-        if: ${{ needs.prepare_chestnut.result == 'success' }}
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        run: |
-          CHESTNUT_BRANCH="${{ needs.prepare_strategy.outputs.new_branch }}-chestnut"
-          CHESTNUT_DIR="${{ github.workspace }}/chestnut_output"
-
-          ${{ env.CI_DIR }}/publish.sh \
-            "${{ github.workspace }}" \
-            "$CHESTNUT_DIR" \
-            "$CHESTNUT_BRANCH" \
-            "${{ needs.prepare_strategy.outputs.version }}" \
-            "https://x-access-token:${{github.token}}@github.com/sunnypilot/sunnypilot.git" \
-            "${{ needs.prepare_strategy.outputs.extra_version_identifier }}"
-
       - name: Tag ${{ needs.prepare_strategy.outputs.environment }}
         if: ${{ needs.prepare_strategy.outputs.is_stable_branch == 'true' && (github.event_name != 'push' || !startsWith(github.ref, 'refs/tags/')) }}
         run: |
@@ -387,12 +465,77 @@ jobs:
             git tag -f -a ${TAG} -m "${{ needs.prepare_strategy.outputs.environment }} @ ${{ needs.prepare_strategy.outputs.version }} of build ${{ needs.prepare_strategy.outputs.build }}."
             git push -f origin ${TAG}
 
+  publish_chestnut:
+    concurrency:
+      group: ${{ needs.prepare_strategy.outputs.publish_concurrency_group }}-chestnut
+      cancel-in-progress: ${{ needs.prepare_strategy.outputs.cancel_publish_in_progress == 'true' }}
+    if: ${{
+      always() && !cancelled() &&
+      needs.build.result == 'success' &&
+      needs.prepare_strategy.result == 'success' &&
+      needs.prepare_small_model.result == 'success' &&
+      needs.prepare_dm_model.result == 'success' &&
+      needs.prepare_chestnut.result == 'success' &&
+      (!contains(github.event_name, 'pull_request') || (github.event.action == 'labeled' && github.event.label.name == 'prebuilt'))
+      }}
+    needs: [ build, prepare_strategy, prepare_chestnut, prepare_small_model, prepare_dm_model ]
+    runs-on: ubuntu-24.04
+    environment: ${{ needs.prepare_strategy.outputs.environment }}
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Download prebuilt artifact
+        uses: actions/download-artifact@v4
+        with:
+          name: prebuilt
+
+      - name: Untar prebuilt
+        run: |
+          mkdir -p ${{ env.OUTPUT_DIR }}
+          tar xzf prebuilt.tar.gz -C ${{ env.OUTPUT_DIR }}
+
+      - name: Download model chunks from HF
+        uses: ./.github/workflows/download-hf-model-chunks
+        with:
+          hf_repo: sunnypilot/sunnypilot_models_v1
+          dest_dir: ${{ env.OUTPUT_DIR }}/openpilot/selfdrive/modeld/models
+          models: |
+            [
+              {"hf_path": "models/defaults/small", "onnx_hash": "${{ needs.prepare_small_model.outputs.driving_onnx_sha256 }}", "canonical": "driving_tinygrad.pkl"},
+              {"hf_path": "models/defaults/dm", "onnx_hash": "${{ needs.prepare_dm_model.outputs.dm_onnx_sha256 }}", "canonical": "dmonitoring_model_tinygrad.pkl"},
+              {"hf_path": "models/defaults/big", "onnx_hash": "${{ needs.prepare_chestnut.outputs.onnx_sha256 }}", "canonical": "big_driving_tinygrad.pkl"}
+            ]
+
+      - name: Configure Git
+        run: |
+          git config --global user.email "github-actions[bot]@users.noreply.github.com"
+          git config --global user.name "github-actions[bot]"
+
+      - name: Publish chestnut branch
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          CHESTNUT_BRANCH="${{ needs.prepare_strategy.outputs.new_branch }}-chestnut"
+
+          ${{ env.CI_DIR }}/publish.sh \
+            "${{ github.workspace }}" \
+            "${{ env.OUTPUT_DIR }}" \
+            "$CHESTNUT_BRANCH" \
+            "${{ needs.prepare_strategy.outputs.version }}" \
+            "https://x-access-token:${{github.token}}@github.com/sunnypilot/sunnypilot.git" \
+            "${{ needs.prepare_strategy.outputs.extra_version_identifier }}"
+
   notify:
     needs:
     - prepare_strategy
     - build
     - publish
+    - publish_chestnut
     - prepare_chestnut
+    - prepare_small_model
+    - prepare_dm_model
     runs-on: ubuntu-24.04
     if: ${{ (always() && !cancelled() && !failure())
       && needs.publish.result == 'success'
@@ -400,6 +543,8 @@ jobs:
       && (fromJSON(vars.DEV_FEEDBACK_NOTIFICATION_BRANCHES_V2)[github.head_ref || github.ref_name] != null) }}
     steps:
       - uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
 
       - name: Prepare notification message
         id: message

openpilot/common/params_keys.h

@@ -195,9 +195,10 @@ inline static std::unordered_map<std::string, ParamKeyAttributes> keys = {
 
     // Model Manager params
     {"ModelManager_ActiveBundle", {PERSISTENT, JSON}},
-    {"ModelManager_ActiveJson", {CLEAR_ON_MANAGER_START, STRING}},
+    {"ModelManager_ActiveBundleUSBGPU", {PERSISTENT, JSON}},
+    {"ModelManager_ActiveJson", {CLEAR_ON_MANAGER_START, JSON}},
     {"ModelManager_ClearCache", {CLEAR_ON_MANAGER_START, BOOL}},
-    {"ModelManager_DownloadIndex", {CLEAR_ON_MANAGER_START | CLEAR_ON_ONROAD_TRANSITION, INT}},
+    {"ModelManager_DownloadRef", {CLEAR_ON_MANAGER_START | CLEAR_ON_ONROAD_TRANSITION, STRING}},
     {"ModelManager_Favs", {PERSISTENT | BACKUP, STRING}},
     {"ModelManager_LastSyncTime", {CLEAR_ON_MANAGER_START | CLEAR_ON_OFFROAD_TRANSITION, INT, "0"}},
     {"ModelManager_LastSyncTime_USBGPU", {CLEAR_ON_MANAGER_START | CLEAR_ON_OFFROAD_TRANSITION, INT, "0"}},

openpilot/selfdrive/modeld/SConscript

@@ -73,44 +73,45 @@ compile_modeld_script = [
 model_w, model_h = MEDMODEL_INPUT_SIZE
 frame_skip = ModelConstants.MODEL_RUN_FREQ // ModelConstants.MODEL_CONTEXT_FREQ
 
-for usbgpu in [False, True] if USBGPU else [False]:
-  target_pkl_path = File(modeld_pkl_path(usbgpu)).abspath
-  # BIG_INTO_SMALL=1 builds the default target from the big model, e.g. to test it without a USB GPU
-  file_prefix, cmd_flags = ('big_', usbgpu_tg_flags) if usbgpu else ('big_' if os.getenv('BIG_INTO_SMALL') else '', tg_flags)
-  driving_onnx_deps = get_existing_chunks(File(f"models/{file_prefix}driving_supercombo.onnx").abspath)
-  camera_res_args = ' '.join(f'{cw}x{ch}' for cw, ch in CAMERA_CONFIGS)
-  # CPU 7 is isolated with isolcpus on AGNOS, so explicitly pin the compiler to it.
-  taskset = 'taskset -c 7 ' if arch == 'comma_arm64' else ''
-  cmd = (f'{cmd_flags} {mac_brew_string} {taskset}python3 {modeld_dir}/compile_modeld.py '
-        f'--model-size {model_w}x{model_h} '
-        f'--camera-resolutions {camera_res_args} '
-        f'--onnx {File(f"models/{file_prefix}driving_supercombo.onnx").abspath} '
-        f'--output {target_pkl_path} --frame-skip {frame_skip}')
-  onnx_sizes_sum = sum(os.path.getsize(f) for f in driving_onnx_deps)
-  chunk_targets = get_chunk_targets(target_pkl_path, estimate_pickle_max_size(onnx_sizes_sum))
-  def do_compile(target, source, env, command=cmd, pkl=target_pkl_path, chunks=chunk_targets):
-    from openpilot.system.hardware.chestnut.flash import link_up
-    # chestnut can enumerate before its PCIe link is up due to varying 12V power behavior across cars
-    for _ in range(10):
-      if link_up():
-        break
-      time.sleep(1)
-    else:
-      print("Chestnut not ready, skipping big model build")
-      return
-    if ret := env.Execute(command):
-      return ret
-    chunk_file(pkl, chunks)
-  def do_chunk(target, source, env, pkl=target_pkl_path, chunks=chunk_targets):
-    chunk_file(pkl, chunks)
-  actions = Action(do_compile, "  [USBGPU] $TARGET") if usbgpu else [cmd, Action(do_chunk, "  [CHUNK] $TARGET")]
-  node = lenv.Command(
-    chunk_targets,
-    tinygrad_files + compile_modeld_script + driving_onnx_deps + [Value(chunk_targets), chunker_file],
-    actions,
-  )
-  if usbgpu:
-    lenv.SideEffect(usbgpu_lock, node)
+if not os.getenv('SKIP_TINYGRAD_COMPILE'):
+  for usbgpu in [False, True] if USBGPU else [False]:
+    target_pkl_path = File(modeld_pkl_path(usbgpu)).abspath
+    # BIG_INTO_SMALL=1 builds the default target from the big model, e.g. to test it without a USB GPU
+    file_prefix, cmd_flags = ('big_', usbgpu_tg_flags) if usbgpu else ('big_' if os.getenv('BIG_INTO_SMALL') else '', tg_flags)
+    driving_onnx_deps = get_existing_chunks(File(f"models/{file_prefix}driving_supercombo.onnx").abspath)
+    camera_res_args = ' '.join(f'{cw}x{ch}' for cw, ch in CAMERA_CONFIGS)
+    # CPU 7 is isolated with isolcpus on AGNOS, so explicitly pin the compiler to it.
+    taskset = 'taskset -c 7 ' if arch == 'comma_arm64' else ''
+    cmd = (f'{cmd_flags} {mac_brew_string} {taskset}python3 {modeld_dir}/compile_modeld.py '
+          f'--model-size {model_w}x{model_h} '
+          f'--camera-resolutions {camera_res_args} '
+          f'--onnx {File(f"models/{file_prefix}driving_supercombo.onnx").abspath} '
+          f'--output {target_pkl_path} --frame-skip {frame_skip}')
+    onnx_sizes_sum = sum(os.path.getsize(f) for f in driving_onnx_deps)
+    chunk_targets = get_chunk_targets(target_pkl_path, estimate_pickle_max_size(onnx_sizes_sum))
+    def do_compile(target, source, env, command=cmd, pkl=target_pkl_path, chunks=chunk_targets):
+      from openpilot.system.hardware.chestnut.flash import link_up
+      # chestnut can enumerate before its PCIe link is up due to varying 12V power behavior across cars
+      for _ in range(10):
+        if link_up():
+          break
+        time.sleep(1)
+      else:
+        print("Chestnut not ready, skipping big model build")
+        return
+      if ret := env.Execute(command):
+        return ret
+      chunk_file(pkl, chunks)
+    def do_chunk(target, source, env, pkl=target_pkl_path, chunks=chunk_targets):
+      chunk_file(pkl, chunks)
+    actions = Action(do_compile, "  [USBGPU] $TARGET") if usbgpu else [cmd, Action(do_chunk, "  [CHUNK] $TARGET")]
+    node = lenv.Command(
+      chunk_targets,
+      tinygrad_files + compile_modeld_script + driving_onnx_deps + [Value(chunk_targets), chunker_file],
+      actions,
+    )
+    if usbgpu:
+      lenv.SideEffect(usbgpu_lock, node)
 
 # get model metadata
 fn = File(f"models/dmonitoring_model").abspath
@@ -142,4 +143,5 @@ def tg_compile(flags, model_name):
      Action(do_chunk, "  [CHUNK] $TARGET")],
   )
 
-tg_compile(tg_flags, 'dmonitoring_model')
+if not os.getenv('SKIP_TINYGRAD_COMPILE'):
+  tg_compile(tg_flags, 'dmonitoring_model')

openpilot/selfdrive/ui/layouts/sidebar.py

@@ -168,9 +168,16 @@ class Sidebar(Widget, SidebarSP):
     # Home/Flag button
     flag_pressed = mouse_down and rl.check_collision_point_rec(mouse_pos, HOME_BTN)
     button_img = self._flag_img if ui_state.started else self._home_img
+    button_pos = rl.Vector2(HOME_BTN.x, HOME_BTN.y)
+    icon_opacity = 1.0
+
+    if gui_app.sunnypilot_ui():
+      button_img, button_pos, icon_opacity = SidebarSP._get_home_icon(self, button_img)
 
     tint = Colors.BUTTON_PRESSED if (ui_state.started and flag_pressed) else Colors.BUTTON_NORMAL
-    rl.draw_texture_ex(button_img, rl.Vector2(HOME_BTN.x, HOME_BTN.y), 0.0, 1.0, tint)
+    if icon_opacity < 1.0:
+      tint = rl.Color(tint[0], tint[1], tint[2], int(255 * icon_opacity))
+    rl.draw_texture_ex(button_img, button_pos, 0.0, 1.0, tint)
 
     # Microphone button
     if self._recording_audio:

openpilot/selfdrive/ui/mici/layouts/home.py

@@ -248,8 +248,11 @@ class MiciHomeLayout(Widget):
 
     # ***** Center-aligned bottom section icons *****
     self._experimental_icon.set_visible(ui_state.experimental_mode)
-    self._egpu_icon.set_visible(ui_state.sm["deviceState"].chestnutPresent and ui_state.usbgpu_compiled)
-    self._egpu_icon_gray.set_visible(ui_state.sm["deviceState"].chestnutPresent and not ui_state.usbgpu_compiled)
+    if gui_app.sunnypilot_ui():
+      self._set_egpu_visibility()
+    else:
+      self._egpu_icon.set_visible(ui_state.sm["deviceState"].chestnutPresent and ui_state.usbgpu_compiled)
+      self._egpu_icon_gray.set_visible(ui_state.sm["deviceState"].chestnutPresent and not ui_state.usbgpu_compiled)
     self._mic_icon.set_visible(ui_state.recording_audio)
     self._body_icon.set_visible(bool(ui_state.is_body))
 

openpilot/selfdrive/ui/sunnypilot/layouts/settings/models.py

@@ -11,6 +11,8 @@ import pyray as rl
 
 from openpilot.cereal import custom
 from openpilot.sunnypilot.models.default_model import get_default_model
+from openpilot.sunnypilot.models.fetcher import ModelFetcher
+from openpilot.sunnypilot.models.helpers import ACTIVE_BUNDLE_KEYS
 from openpilot.common.constants import CV
 from openpilot.selfdrive.ui.ui_state import device, ui_state
 from openpilot.system.ui.lib.multilang import tr
@@ -68,7 +70,7 @@ class ModelsLayout(Widget):
       callback=self._clear_cache
     )
 
-    self.cancel_download_item = button_item(tr("Cancel Download"), tr("Cancel"), "", lambda: ui_state.params.remove("ModelManager_DownloadIndex"))
+    self.cancel_download_item = button_item(tr("Cancel Download"), tr("Cancel"), "", lambda: ui_state.params.remove("ModelManager_DownloadRef"))
 
     self.lane_turn_value_control = option_item_sp(tr("Adjust Lane Turn Speed"), "LaneTurnValue", 500, 2000,
                                                   tr("Set the maximum speed for lane turn desires. Default is 19 mph."),
@@ -115,8 +117,12 @@ class ModelsLayout(Widget):
   def calculate_cache_size():
     cache_size = 0.0
     if os.path.exists(CUSTOM_MODEL_PATH):
-      cache_size = sum(os.path.getsize(os.path.join(CUSTOM_MODEL_PATH, file)) for file in os.listdir(CUSTOM_MODEL_PATH)) / (1024**2)
-    return cache_size
+      for file in os.listdir(CUSTOM_MODEL_PATH):
+        try:
+          cache_size += os.path.getsize(os.path.join(CUSTOM_MODEL_PATH, file))
+        except OSError:
+          continue
+    return cache_size / (1024**2)
 
   def _clear_cache(self):
     def _callback(response):
@@ -142,7 +148,7 @@ class ModelsLayout(Widget):
     if not bundle:
       return
 
-    self.cancel_download_item.set_visible(bool(self.model_manager.selectedBundle) and ui_state.params.get("ModelManager_DownloadIndex") is not None)
+    self.cancel_download_item.set_visible(bool(self.model_manager.selectedBundle) and ui_state.params.get("ModelManager_DownloadRef") is not None)
 
     if (current_time := time.monotonic()) - self.last_cache_calc_time > 0.5:
       self.last_cache_calc_time = current_time
@@ -183,9 +189,10 @@ class ModelsLayout(Widget):
       return
     selected_ref = self.model_dialog.selection_ref
     if selected_ref == "Default":
-      ui_state.params.remove("ModelManager_ActiveBundle")
+      source = ModelFetcher.active_source(ui_state.sm["deviceState"].chestnutPresent)
+      ui_state.params.remove(ACTIVE_BUNDLE_KEYS[source])
     elif selected_bundle := next((bundle for bundle in self.model_manager.availableBundles if bundle.ref == selected_ref), None):
-      ui_state.params.put("ModelManager_DownloadIndex", selected_bundle.index)
+      ui_state.params.put("ModelManager_DownloadRef", selected_bundle.ref)
     self.model_dialog = None
 
   @staticmethod
@@ -223,7 +230,7 @@ class ModelsLayout(Widget):
     advanced_controls: bool = ui_state.params.get_bool("ShowAdvancedControls")
     turn_desire: bool = ui_state.params.get_bool("LaneTurnDesire")
     live_delay: bool = ui_state.params.get_bool("LagdToggle")
-    camera_offset: bool = ui_state.params.get("ModelManager_ActiveBundle") is not None
+    camera_offset: bool = ui_state.active_bundle is not None
 
     self.lane_turn_desire_toggle.action_item.set_state(turn_desire)
     self.lane_turn_value_control.set_visible(turn_desire and advanced_controls)
@@ -237,8 +244,9 @@ class ModelsLayout(Widget):
     self._update_lagd_description(live_delay)
     self.model_manager = ui_state.sm["modelManagerSP"]
     self._handle_bundle_download_progress()
-    default_label = f"{get_default_model()} (Default)"
-    active_name = self.model_manager.activeBundle.displayName if self.model_manager and self.model_manager.activeBundle.ref else default_label
+    # read the slot through ui_state, not modelManagerSP: the manager republishes a
+    # tick after a chestnut change, and the stale bundle flashes the wrong model
+    active_name = (ui_state.active_bundle or {}).get("displayName") or f"{get_default_model()} (Default)"
     self.current_model_item.action_item.set_value(active_name)
 
     if not ui_state.is_offroad():

openpilot/selfdrive/ui/sunnypilot/layouts/sidebar.py

@@ -4,11 +4,14 @@ Copyright (c) 2021-, Haibin Wen, sunnypilot, and a number of other contributors.
 This file is part of sunnypilot and is licensed under the MIT License.
 See the LICENSE.md file in the root directory for more details.
 """
+import math
+
 import pyray as rl
 import time
 from dataclasses import dataclass
 from openpilot.selfdrive.ui.ui_state import ui_state
 from openpilot.sunnypilot.sunnylink.api import UNREGISTERED_SUNNYLINK_DONGLE_ID
+from openpilot.system.ui.lib.application import gui_app
 from openpilot.system.ui.lib.multilang import tr_noop
 
 
@@ -18,6 +21,9 @@ METRIC_MARGIN = 30
 METRIC_START_Y = 300
 HOME_BTN = rl.Rectangle(60, 860, 180, 180)
 
+EGPU_ICON_WIDTH = 180
+EGPU_ICON_HEIGHT = 133
+
 
 # Color scheme
 class Colors:
@@ -53,6 +59,10 @@ class MetricData:
 class SidebarSP:
   def __init__(self):
     self._sunnylink_status = MetricData(tr_noop("SUNNYLINK"), tr_noop("OFFLINE"), Colors.WARNING)
+    self._egpu_green_img = gui_app.texture("icons_mici/egpu_green.png", EGPU_ICON_WIDTH, EGPU_ICON_HEIGHT)
+    self._egpu_default_img = gui_app.texture("icons_mici/egpu.png", EGPU_ICON_WIDTH, EGPU_ICON_HEIGHT)
+    self._egpu_orange_img = gui_app.texture("icons_mici/egpu_orange.png", EGPU_ICON_WIDTH, EGPU_ICON_HEIGHT)
+    self._egpu_gray_img = gui_app.texture("icons_mici/egpu_gray.png", EGPU_ICON_WIDTH, EGPU_ICON_HEIGHT)
 
   def _update_sunnylink_status(self):
     if not ui_state.params.get_bool("SunnylinkEnabled"):
@@ -78,6 +88,29 @@ class SidebarSP:
 
     self._sunnylink_status.update(tr_noop("SUNNYLINK"), status, color)
 
+  def _get_home_icon(self, default_img: rl.Texture) -> tuple[rl.Texture, rl.Vector2, float]:
+    default_pos = rl.Vector2(HOME_BTN.x, HOME_BTN.y)
+    if not ui_state.sm["deviceState"].chestnutPresent:
+      return default_img, default_pos, 1.0
+
+    big_model_selected = ui_state.usbgpu_compiled or ui_state.model_runner_tinygrad
+    big_model_failed = ui_state.started and ui_state.big_model_failed
+    loading = ui_state.usbgpu_loading or (big_model_selected and ui_state.started and ui_state.usbgpu_active is None)
+
+    if loading:
+      icon = self._egpu_default_img
+      opacity = 0.35 + 0.65 * (0.5 - 0.5 * math.cos(rl.get_time() * 6.0))
+    elif big_model_selected and big_model_failed:
+      icon, opacity = self._egpu_orange_img, 1.0
+    elif big_model_selected:
+      icon, opacity = self._egpu_green_img, 1.0
+    else:
+      icon, opacity = self._egpu_gray_img, 1.0
+
+    x = HOME_BTN.x + (HOME_BTN.width - icon.width) / 2
+    y = HOME_BTN.y + (HOME_BTN.height - icon.height) / 2
+    return icon, rl.Vector2(x, y), opacity
+
   def _draw_metrics_w_sunnylink(self, rect: rl.Rectangle, _temp, _panda, _connect):
     metrics = [_temp, _panda, _connect, self._sunnylink_status]
     start_y = int(rect.y) + METRIC_START_Y

openpilot/selfdrive/ui/sunnypilot/mici/layouts/home.py

@@ -4,8 +4,14 @@ Copyright (c) 2021-, Haibin Wen, sunnypilot, and a number of other contributors.
 This file is part of sunnypilot and is licensed under the MIT License.
 See the LICENSE.md file in the root directory for more details.
 """
+import math
+
+import pyray as rl
+
 from openpilot.selfdrive.ui.mici.layouts.home import MiciHomeLayout
+from openpilot.selfdrive.ui.ui_state import ui_state
 from openpilot.system.ui.lib.application import FontWeight
+from openpilot.system.ui.widgets.icon_widget import IconWidget
 from openpilot.system.ui.widgets.label import UnifiedLabel
 
 
@@ -13,3 +19,35 @@ class MiciHomeLayoutSP(MiciHomeLayout):
   def __init__(self):
     super().__init__()
     self._openpilot_label = UnifiedLabel("sunnypilot", font_size=88, font_weight=FontWeight.AUDIOWIDE, max_width=480, wrap_text=False)
+    self._egpu_icon_default = IconWidget("icons_mici/egpu.png", (50, 37))
+    self._egpu_icon_default.set_visible(False)
+    self._egpu_icon_orange = IconWidget("icons_mici/egpu_orange.png", (50, 37))
+    self._egpu_icon_orange.set_visible(False)
+    gray_idx = self._status_bar_layout.widgets.index(self._egpu_icon_gray)
+    self._status_bar_layout.widgets.insert(gray_idx + 1, self._egpu_icon_default)
+    self._status_bar_layout.widgets.insert(gray_idx + 2, self._egpu_icon_orange)
+
+  def _set_egpu_visibility(self):
+    chestnut = ui_state.sm["deviceState"].chestnutPresent
+    if not chestnut:
+      self._egpu_icon.set_visible(False)
+      self._egpu_icon_default.set_visible(False)
+      self._egpu_icon_orange.set_visible(False)
+      self._egpu_icon_gray.set_visible(False)
+      return
+
+    big_model_selected = ui_state.usbgpu_compiled or ui_state.model_runner_tinygrad
+    big_model_failed = ui_state.started and ui_state.big_model_failed
+    loading = ui_state.usbgpu_loading or (big_model_selected and ui_state.started and ui_state.usbgpu_active is None)
+
+    if loading:
+      self._egpu_icon_default._opacity = 0.35 + 0.65 * (0.5 - 0.5 * math.cos(rl.get_time() * 6.0))
+      self._egpu_icon_default.set_visible(True)
+      self._egpu_icon.set_visible(False)
+      self._egpu_icon_orange.set_visible(False)
+      self._egpu_icon_gray.set_visible(False)
+    else:
+      self._egpu_icon_default.set_visible(False)
+      self._egpu_icon.set_visible(big_model_selected and not big_model_failed)
+      self._egpu_icon_orange.set_visible(big_model_selected and big_model_failed)
+      self._egpu_icon_gray.set_visible(not big_model_selected)

openpilot/selfdrive/ui/sunnypilot/mici/layouts/models.py

@@ -8,6 +8,8 @@ import pyray as rl
 
 from openpilot.cereal import custom
 from openpilot.sunnypilot.models.default_model import get_default_model
+from openpilot.sunnypilot.models.fetcher import ModelFetcher
+from openpilot.sunnypilot.models.helpers import ACTIVE_BUNDLE_KEYS
 from openpilot.selfdrive.ui.mici.widgets.button import BigButton
 from openpilot.selfdrive.ui.sunnypilot.layouts.settings.models import ModelsLayout
 from openpilot.selfdrive.ui.ui_state import ui_state, device
@@ -60,7 +62,7 @@ class ModelsLayoutMici(NavScroller):
     self.select_model_btn.set_click_callback(self._show_folders)
 
     self.cancel_download_btn = BigButton(tr("cancel download"))
-    self.cancel_download_btn.set_click_callback(lambda: ui_state.params.remove("ModelManager_DownloadIndex"))
+    self.cancel_download_btn.set_click_callback(lambda: ui_state.params.remove("ModelManager_DownloadRef"))
 
     self.main_items = [self.current_model_info, self.select_model_btn, self.cancel_download_btn]
     self._scroller.add_widgets(self.main_items)
@@ -113,11 +115,12 @@ class ModelsLayoutMici(NavScroller):
     gui_app.pop_widgets_to(self)
 
   def _select_model(self, bundle):
-    ui_state.params.put("ModelManager_DownloadIndex", bundle.index)
+    ui_state.params.put("ModelManager_DownloadRef", bundle.ref)
     self._pop_to_main()
 
   def _select_default(self):
-    ui_state.params.remove("ModelManager_ActiveBundle")
+    source = ModelFetcher.active_source(ui_state.sm["deviceState"].chestnutPresent)
+    ui_state.params.remove(ACTIVE_BUNDLE_KEYS[source])
     self._pop_to_main()
 
   def _select_folder(self, folder_name):
@@ -162,8 +165,9 @@ class ModelsLayoutMici(NavScroller):
     self._was_downloading = is_downloading
 
     self.current_model_info.current_model_header.set_text(tr("active model"))
-    default_model_text = f"{get_default_model()} (Default)".lower()
-    model_text = manager.activeBundle.displayName.lower() if manager.activeBundle.ref else default_model_text
+    # read the slot through ui_state, not modelManagerSP: the manager republishes a
+    # tick after a chestnut change, and the stale bundle flashes the wrong model
+    model_text = ((ui_state.active_bundle or {}).get("displayName") or f"{get_default_model()} (Default)").lower()
     self.current_model_info.current_model_text.set_text(model_text)
     self.current_model_info.info_header.set_text(tr("cache size"))
     self.current_model_info.info_text.set_text(f"{ModelsLayout.calculate_cache_size():.2f} MB")

openpilot/selfdrive/ui/sunnypilot/mici/onroad/hud_renderer.py

@@ -7,6 +7,7 @@ See the LICENSE.md file in the root directory for more details.
 import pyray as rl
 
 from openpilot.selfdrive.ui.mici.onroad.hud_renderer import HudRenderer
+from openpilot.selfdrive.ui.ui_state import ui_state
 from openpilot.selfdrive.ui.sunnypilot.onroad.blind_spot_indicators import BlindSpotIndicators
 
 
@@ -21,6 +22,8 @@ class HudRendererSP(HudRenderer):
 
   def _render(self, rect: rl.Rectangle) -> None:
     super()._render(rect)
+    if ui_state.usbgpu and not ui_state.usbgpu_compiled and ui_state.model_runner_tinygrad:
+      self._draw_model_source(rect)
     self.blind_spot_indicators.render(rect)
 
   def _has_blind_spot_detected(self) -> bool:

openpilot/selfdrive/ui/sunnypilot/ui_state.py

@@ -10,6 +10,7 @@ from openpilot.cereal import messaging, log, custom
 from opendbc.car.structs import car
 from openpilot.common.params import Params
 from openpilot.selfdrive.ui.sunnypilot.layouts.settings.display import OnroadBrightness
+from openpilot.sunnypilot.models.helpers import ACTIVE_BUNDLE_KEYS, get_active_source
 from openpilot.sunnypilot.sunnylink.sunnylink_state import SunnylinkState
 from openpilot.system.ui.lib.application import gui_app
 from openpilot.system.ui.sunnypilot.widgets.screen_saver import ScreenSaverSP
@@ -43,6 +44,7 @@ class UIStateSP:
     self.screensaver_enabled: bool = False
 
     self.active_bundle = None
+    self.model_runner_tinygrad: bool = False
     self.blindspot: bool = False
     self.chevron_metrics = None
     self.custom_interactive_timeout: int = 0
@@ -150,7 +152,10 @@ class UIStateSP:
       self.has_icbm = self.CP_SP.intelligentCruiseButtonManagementAvailable and self.params.get_bool("IntelligentCruiseButtonManagement")
 
     self._enforce_constraints()
-    self.active_bundle = self.params.get("ModelManager_ActiveBundle")
+    source = get_active_source(usbgpu=self.usbgpu, usbgpu_active=self.usbgpu_active,
+                               usbgpu_loading=self.usbgpu_loading, offroad=self.is_offroad())
+    self.active_bundle = self.params.get(ACTIVE_BUNDLE_KEYS[source])
+    self.model_runner_tinygrad = self.active_bundle is not None and self.active_bundle.get("runner") == "tinygrad"
     self.blindspot = self.params.get_bool("BlindSpot")
     self.chevron_metrics = self.params.get("ChevronInfo")
     self.custom_interactive_timeout = self.params.get("InteractivityTimeout", return_default=True)

openpilot/selfdrive/ui/ui_state.py

@@ -112,6 +112,15 @@ class UIState(UIStateSP):
   def add_on_body_changed_callbacks(self, callback: Callable[[], None]):
     self._on_body_changed_callbacks.append(callback)
 
+  @property
+  def big_model_failed(self) -> bool:
+    # Mirrors the onroad HUD's four-condition check so sidebar and home icons reflect the same failure states
+    return (self.usbgpu_active is False or
+            not self.sm['deviceState'].chestnutPresent or
+            (self.usbgpu_active is True and self.sm.recv_frame['modelV2'] > self.started_frame and
+             not self.sm.alive['modelV2']) or
+            (self.usbgpu_active is None and self.sm.recv_frame['modelV2'] > self.started_frame))
+
   @property
   def engaged(self) -> bool:
     return self.started and (self.sm["selfdriveState"].enabled or self.sm["selfdriveStateSP"].mads.enabled)

openpilot/sunnypilot/modeld_v2/compile_modeld.py

@@ -7,6 +7,7 @@ See the LICENSE.md file in the root directory for more details.
 """
 
 import argparse
+import math
 import os
 import tempfile
 import time
@@ -66,14 +67,15 @@ def get_policy_npy_shapes(input_shapes: dict, is_supercombo: bool = False) -> tu
   if desire_key:
     shapes['desire'] = (input_shapes[desire_key][2],)
 
-  if is_supercombo and 'features_buffer' in input_shapes:
-    fb = input_shapes['features_buffer']
-    shapes['prev_feat'] = (fb[0], fb[2])
-
   for key, shape in input_shapes.items():
     if key not in (desire_key, 'features_buffer') and 'img' not in key:
       shapes[key] = tuple(shape)
 
+  if is_supercombo and 'features_buffer' in input_shapes:
+    fb = input_shapes['features_buffer']
+    feat_dim = math.prod(fb[2:])
+    shapes['prev_feat'] = (fb[0], feat_dim)
+
   sizes = [int(np.prod(size)) for size in shapes.values()]
   return shapes, sizes
 
@@ -117,8 +119,9 @@ def generate_queues_and_npy(input_shapes: dict, frame_skip: int, device: str = D
   }
 
   if features_buffer:
+    feat_dim = math.prod(features_buffer[2:])
     feat_q_len = frame_skip * features_buffer[1] if is_supercombo else frame_skip * (features_buffer[1] - 1) + 1
-    queues['feat_q'] = Tensor(np.zeros((feat_q_len, features_buffer[0], features_buffer[2]),
+    queues['feat_q'] = Tensor(np.zeros((feat_q_len, features_buffer[0], feat_dim),
                        dtype=np.float32), device=device).contiguous().realize()
 
   queues.update({key: Tensor(value, device='NPY').realize() for key, value in npy_arrays.items() if key in ('tfm', 'big_tfm')})
@@ -183,14 +186,14 @@ def make_run_policy(vision_runner, policy_runners: list, features_slice: slice,
     warped_dev = warped.to(Device.DEFAULT)
     Tensor.realize(packed_npy_inputs_dev, warped_dev)
 
-    img = shift_and_sample(img_q, warped_dev[0:1], sample_skip_fn).realize()
-    big_img = shift_and_sample(big_img_q, warped_dev[1:2], sample_skip_fn).realize()
+    img = shift_and_sample(img_q, warped_dev[0:1], sample_skip_fn)
+    big_img = shift_and_sample(big_img_q, warped_dev[1:2], sample_skip_fn)
 
     unpacked_tensors = [tensor.reshape(shape) for tensor, shape in zip(packed_npy_inputs_dev.split(npy_sizes), npy_shapes.values(), strict=True)]
     unpacked_dict = dict(zip(npy_shapes.keys(), unpacked_tensors, strict=True))
 
     desire_dev = unpacked_dict['desire']
-    desire_buf = shift_and_sample(desire_q, desire_dev.reshape(1, 1, -1), sample_desire_fn).realize()
+    desire_buf = shift_and_sample(desire_q, desire_dev.reshape(1, 1, -1), sample_desire_fn)
 
     inputs = {desire_key: desire_buf}
     for key, tensor_val in unpacked_dict.items():
@@ -199,7 +202,7 @@ def make_run_policy(vision_runner, policy_runners: list, features_slice: slice,
 
     if 'prev_feat' in unpacked_dict:
       prev_feat_dev = unpacked_dict['prev_feat']
-      inputs['features_buffer'] = shift_and_sample(feat_q, prev_feat_dev.reshape(1, 1, -1), sample_skip_fn).realize()
+      inputs['features_buffer'] = shift_and_sample(feat_q, prev_feat_dev.reshape(1, 1, -1), sample_skip_fn).reshape(input_shapes['features_buffer'])
 
     if vision_runner:
       vision_out_cast = next(iter(vision_runner({road_key: img, wide_key: big_img}).values())).cast('float32').realize()
@@ -211,7 +214,7 @@ def make_run_policy(vision_runner, policy_runners: list, features_slice: slice,
 
     inputs.update({road_key: img, wide_key: big_img})
     if 'features_buffer' not in inputs:
-      inputs['features_buffer'] = sample_skip_fn(feat_q)
+      inputs['features_buffer'] = sample_skip_fn(feat_q).reshape(input_shapes['features_buffer'])
 
     policy_out = next(iter(policy_runners[0](inputs).values())).cast('float32').realize()
     if 'features_buffer' not in inputs and features_slice is not None:

openpilot/sunnypilot/modeld_v2/modeld.py

@@ -91,7 +91,7 @@ class ModelState(ModelStateBase):
     if env_pkl and os.path.exists(env_pkl):
       model_bundle = None
     else:
-      model_bundle = get_active_bundle()
+      model_bundle = get_active_bundle(usbgpu=usbgpu)
     self.generation = model_bundle.generation if model_bundle is not None else None
     overrides = {override.key: override.value for override in model_bundle.overrides} if model_bundle else {}
 

openpilot/sunnypilot/modeld_v2/tests/helpers.py

@@ -190,8 +190,8 @@ def tmp_path():
 
 def patch_modeld(monkeypatch):
   def _patch(bundle):
-    monkeypatch.setattr(helpers, 'get_active_bundle', lambda params=None: bundle)
-    monkeypatch.setattr(modeld_module, 'get_active_bundle', lambda params=None: bundle)
+    monkeypatch.setattr(helpers, 'get_active_bundle', lambda params=None, *, usbgpu=None: bundle)
+    monkeypatch.setattr(modeld_module, 'get_active_bundle', lambda params=None, *, usbgpu=None: bundle)
 
   return _patch
 

openpilot/sunnypilot/modeld_v2/tests/test_combined_pkl_loader.py

@@ -59,8 +59,8 @@ class TestFindDrivingPkl(OpenpilotTestCase):
 class TestModelStateCombinedInit(OpenpilotTestCase):
   def test_asserts_when_no_pkl(self, monkeypatch):
     bundle = DummyBundle(models=[], is_20hz=True)
-    monkeypatch.setattr(helpers, 'get_active_bundle', lambda params=None: bundle)
-    monkeypatch.setattr(modeld_module, 'get_active_bundle', lambda params=None: bundle)
+    monkeypatch.setattr(helpers, 'get_active_bundle', lambda params=None, *, usbgpu=None: bundle)
+    monkeypatch.setattr(modeld_module, 'get_active_bundle', lambda params=None, *, usbgpu=None: bundle)
     with self.assertRaisesRegex(AssertionError, "No driving pkl found"):
       ModelState(cam_w=CAM_W, cam_h=CAM_H)
 

openpilot/sunnypilot/modeld_v2/tests/test_compile_modeld.py

@@ -195,3 +195,85 @@ class TestReadFileChunkedToDisk(OpenpilotTestCase):
 
       assert out.parent == Path(d)
       assert out.read_bytes() == payload
+
+
+class Test4DFeaturesBuffer(OpenpilotTestCase):
+  def test_get_policy_npy_shapes_4d(self):
+    from openpilot.sunnypilot.modeld_v2.compile_modeld import get_policy_npy_shapes
+    input_shapes = {
+      'desire_pulse': (1, 25, 8),
+      'features_buffer': (1, 24, 32, 512),  # compare 4d to 3d for regression
+      'traffic_convention': (1, 2),
+      'action_t': (1, 2)
+    }
+    shapes, sizes = get_policy_npy_shapes(input_shapes, is_supercombo=True)
+    assert shapes['prev_feat'] == (1, 16384)
+    assert sizes == [8, 2, 2, 16384]
+
+  def test_get_policy_npy_shapes_3d(self):
+    from openpilot.sunnypilot.modeld_v2.compile_modeld import get_policy_npy_shapes
+    input_shapes = {
+      'desire_pulse': (1, 25, 8),
+      'features_buffer': (1, 24, 512),
+      'traffic_convention': (1, 2),
+      'action_t': (1, 2)
+    }
+    shapes, sizes = get_policy_npy_shapes(input_shapes, is_supercombo=True)
+    assert shapes['prev_feat'] == (1, 512)
+    assert sizes == [8, 2, 2, 512]
+
+
+class TestStockCompileModeldEquivalence(OpenpilotTestCase):
+  def test_get_policy_npy_shapes_matches_stock(self):
+    from openpilot.selfdrive.modeld.compile_modeld import get_policy_npy_shapes as stock_get_policy_npy_shapes
+    from openpilot.sunnypilot.modeld_v2.compile_modeld import get_policy_npy_shapes as sunny_get_policy_npy_shapes
+
+    stock_input_shapes = {
+      'desire_pulse': (1, 25, 8),
+      'features_buffer': (1, 24, 512), # see below comment
+      'traffic_convention': (1, 2),
+      'action_t': (1, 2),
+    }
+
+    stock_shapes, stock_sizes = stock_get_policy_npy_shapes(stock_input_shapes)
+    sunny_shapes, sunny_sizes = sunny_get_policy_npy_shapes(stock_input_shapes, is_supercombo=True)
+
+    assert sunny_shapes == stock_shapes
+    assert sunny_sizes == stock_sizes
+    assert sunny_shapes['prev_feat'] == (1, 512)
+
+  def test_make_input_queues_full_stock_equivalence(self):
+    from openpilot.selfdrive.modeld.compile_modeld import make_input_queues as stock_make_input_queues
+    from openpilot.sunnypilot.modeld_v2.compile_modeld import make_supercombo_input_queues as sunny_make_supercombo_input_queues
+    input_shapes = {
+      'img': (1, 12, 128, 256),
+      'desire_pulse': (1, 25, 8),
+      'features_buffer': (1, 24, 512),  # when https://github.com/commaai/openpilot/pull/38681 merges, update to 1,24,32,512
+      'traffic_convention': (1, 2),
+      'action_t': (1, 2),
+    }
+    frame_skip = 4
+
+    stock_queues, stock_npy = stock_make_input_queues(input_shapes, frame_skip, device='NPY')
+    sunny_queues, sunny_npy = sunny_make_supercombo_input_queues(input_shapes, frame_skip, device='NPY')
+    assert set(sunny_queues.keys()) == set(stock_queues.keys())
+    for key in stock_queues:
+      assert sunny_queues[key].shape == stock_queues[key].shape, \
+        f"Queue shape mismatch for {key}: sunny {sunny_queues[key].shape} != stock {stock_queues[key].shape}"
+    assert set(sunny_npy.keys()) == set(stock_npy.keys())
+    for key in stock_npy:
+      assert sunny_npy[key].shape == stock_npy[key].shape, \
+        f"Numpy array shape mismatch for {key}: sunny {sunny_npy[key].shape} != stock {stock_npy[key].shape}"
+
+  def test_make_warp_queues_stock_equivalence(self):
+    from openpilot.selfdrive.modeld.compile_modeld import make_warp_input_queues as stock_make_warp_queues
+    from openpilot.sunnypilot.modeld_v2.compile_modeld import make_warp_queues as sunny_make_warp_queues
+    stock_vision_shapes = {'img': (1, 12, 128, 256)}  # for now?
+    stock_queues, stock_npy = stock_make_warp_queues(stock_vision_shapes, frame_skip=4, device='NPY')
+    sunny_queues, sunny_npy = sunny_make_warp_queues(device='NPY')
+
+    assert set(sunny_npy.keys()) == set(stock_npy.keys()) == {'tfm', 'big_tfm'}
+    for key in sunny_npy:
+      assert sunny_npy[key].shape == stock_npy[key].shape == (3, 3)
+
+

openpilot/sunnypilot/models/fetcher.py

@@ -13,8 +13,6 @@ from openpilot.common.params import Params
 from openpilot.common.swaglog import cloudlog
 from openpilot.common.hardware.hw import Paths
 from openpilot.sunnypilot.models.helpers import is_bundle_version_compatible
-from openpilot.selfdrive.modeld.helpers import usbgpu_present
-
 from openpilot.cereal import custom
 
 
@@ -140,45 +138,53 @@ class ModelCache:
 
 class ModelFetcher:
   """Handles fetching and caching of model data from remote source"""
-  MODEL_URL = "https://raw.githubusercontent.com/sunnypilot/sunnypilot-models/refs/heads/gh-pages/docs/driving_models_v20.json"
-  MODEL_URL_USBGPU = "https://raw.githubusercontent.com/sunnypilot/sunnypilot-models/refs/heads/gh-pages/docs/driving_models_usbgpu_v21.json"
+  MODEL_URL = "https://raw.githubusercontent.com/sunnypilot/sunnypilot-models/refs/heads/gh-pages/docs/driving_models_v21.json"
+  MODEL_URL_USBGPU = "https://raw.githubusercontent.com/sunnypilot/sunnypilot-models/refs/heads/gh-pages/docs/driving_models_usbgpu_v22.json"
+
+  MODEL_SOURCES = {
+    "qcom": (MODEL_URL, ""),
+    "usbgpu": (MODEL_URL_USBGPU, "_USBGPU"),
+  }
 
   def __init__(self, params: Params):
     self.params = params
     self.model_parser = ModelParser()
-    self._is_usbgpu: bool | None = None
-    self.model_cache = ModelCache(params)
-    self.model_url = self.MODEL_URL
-    self._update_model_source()
+    self.model_caches = {
+      source: ModelCache(params, suffix=suffix)
+      for source, (_, suffix) in self.MODEL_SOURCES.items()
+    }
+    self._refetched: set[str] = set()
+    self.params.put("ModelManager_ActiveJson", {
+      "qcom": self.MODEL_URL,
+      "usbgpu": self.MODEL_URL_USBGPU,
+    }, block=True)
 
-  def _update_model_source(self) -> None:
-    """Updates what json to use based on usbgpu availability"""
-    is_usbgpu = usbgpu_present()
-    if is_usbgpu != self._is_usbgpu:
-      self._is_usbgpu = is_usbgpu
-      self.model_cache = ModelCache(self.params, suffix="_USBGPU" if is_usbgpu else "")
-      self.model_url = self.MODEL_URL_USBGPU if is_usbgpu else self.MODEL_URL
-      self.params.put("ModelManager_ActiveJson", self.model_url, block=True)
+  @staticmethod
+  def active_source(chestnut_present: bool) -> str:
+    return "usbgpu" if chestnut_present else "qcom"
 
-  def _fetch_and_cache_models(self) -> list[custom.ModelManagerSP.ModelBundle] | None:
+  def _fetch_and_cache_models(self, source: str) -> list[custom.ModelManagerSP.ModelBundle] | None:
     """Fetches fresh model data from remote and updates cache.
     Returns None on transport errors. Raises on 404 and other fatal HTTP errors.
     """
+    model_url, _ = self.MODEL_SOURCES[source]
     try:
-      response = requests.get(self.model_url, timeout=10)
+      response = requests.get(model_url, timeout=10)
 
       # Explicitly handle 404 differently
       if response.status_code == 404:
-        cloudlog.error(f"Models URL returned 404 Not Found: {self.model_url}")
-        raise HTTPError(f"404 Not Found: {self.model_url}", response=response)
+        cloudlog.error(f"Models URL returned 404 Not Found: {model_url}")
+        raise HTTPError(f"404 Not Found: {model_url}", response=response)
 
       # Raise for any other 4xx/5xx
       response.raise_for_status()
 
       json_data = response.json()
-      self.model_cache.set(json_data)
-      cloudlog.debug("Successfully updated models cache")
-      return self.model_parser.parse_models(json_data)
+      parsed = self.model_parser.parse_models(json_data)
+      if parsed:
+        self.model_caches[source].set(json_data)
+        cloudlog.debug(f"Successfully updated models cache for {source}")
+      return parsed
 
     except ConnectionError as e:
       cloudlog.warning(f"DNS/connection error while fetching models: {e}")
@@ -191,16 +197,40 @@ class ModelFetcher:
 
     return None
 
-  def get_available_bundles(self) -> list[custom.ModelManagerSP.ModelBundle]:
-    """Gets the list of available models, with smart cache handling"""
-    self._update_model_source()
-    cached_data, is_expired = self.model_cache.get()
+  @staticmethod
+  def _cache_matches_source(source: str, cached_data: dict) -> bool:
+    bundles = cached_data.get("bundles", [])
+    if source == "usbgpu":
+      return any(bundle.get("is_big") is True for bundle in bundles)
+    return not any(bundle.get("is_big") is True for bundle in bundles)
+
+  def get_bundles_for_source(self, source: str) -> list[custom.ModelManagerSP.ModelBundle]:
+    if source not in self.MODEL_SOURCES:
+      cloudlog.warning(f"Unknown model source: {source}")
+      return []
+
+    cached_data, is_expired = self.model_caches[source].get()
 
     if cached_data and not is_expired:
-      cloudlog.debug("Using valid cached models data")
-      return self.model_parser.parse_models(cached_data)
+      # a source is refetched over a mismatch at most once per process: if the fresh
+      # manifest still mismatches, the URL is authoritative and the cache is trusted
+      if self._cache_matches_source(source, cached_data) or source in self._refetched:
+        try:
+          parsed = self.model_parser.parse_models(cached_data)
+        except Exception:
+          cloudlog.warning(f"Failed to parse cached models for {source}; refetching", exc_info=True)
+        else:
+          if parsed:
+            cloudlog.debug(f"Using valid cached models data for source {source}")
+            return parsed
+          # a source-matching cache that yields no valid bundles is stale (e.g. an old
+          # manifest version) - do not trust it, refetch so the source is repopulated
+          cloudlog.warning(f"Cached models for {source} have no valid bundles; refetching")
+      else:
+        self._refetched.add(source)
+        cloudlog.warning(f"Cached models for {source} not valid; refetching once")
 
-    fetched_bundles = self._fetch_and_cache_models()
+    fetched_bundles = self._fetch_and_cache_models(source)
     if fetched_bundles is not None:
       return fetched_bundles
 
@@ -208,12 +238,33 @@ class ModelFetcher:
       cloudlog.warning("Failed to fetch fresh data and no cache available")
 
     cloudlog.warning("Failed to fetch fresh data. Using expired cache as fallback")
-    return self.model_parser.parse_models(cached_data)
+    try:
+      return self.model_parser.parse_models(cached_data)
+    except Exception:
+      return []
+
+
+def get_cached_bundles(params: Params, source: str) -> list[custom.ModelManagerSP.ModelBundle]:
+
+  if source not in ModelFetcher.MODEL_SOURCES:
+    cloudlog.warning(f"Unknown model source: {source}")
+    return []
+  _, suffix = ModelFetcher.MODEL_SOURCES[source]
+  cached_data = params.get(f"ModelManager_ModelsCache{suffix}")
+  if not cached_data:
+    return []
+  try:
+    return ModelParser.parse_models(cached_data)
+  except Exception as e:
+    cloudlog.warning(f"Failed to parse cached models for source {source}: {e}")
+    return []
+
 
 if __name__ == "__main__":
+  from openpilot.selfdrive.modeld.helpers import usbgpu_present
   params = Params()
   model_fetcher = ModelFetcher(params)
-  bundles = model_fetcher.get_available_bundles()
+  bundles = model_fetcher.get_bundles_for_source(ModelFetcher.active_source(usbgpu_present()))
   for bundle in bundles:
     for model in bundle.models:
       model_overrides = {override.key: override.value for override in bundle.overrides}

openpilot/sunnypilot/models/helpers.py

@@ -16,14 +16,20 @@ from openpilot.common.params import Params
 from openpilot.common.swaglog import cloudlog
 from openpilot.sunnypilot.models.constants import Meta, MetaSimPose, MetaTombRaider
 from openpilot.common.hardware.hw import Paths
+from openpilot.selfdrive.modeld.helpers import usbgpu_present
 
 # SET ME TO THE EXACT JSON VERSION WE SET IN SUNNYPILOT_MODELS REPO
-REQUIRED_JSON_VERSION = 17
+REQUIRED_JSON_VERSION = 18
 
 CUSTOM_MODEL_PATH = Paths.model_root()
 METADATA_PATH = Path(__file__).parent / '../models/supercombo_metadata.pkl'
 ModelManager = custom.ModelManagerSP
-_LAST_VALIDATED_RAW = None
+
+ACTIVE_BUNDLE_KEYS = {
+  "qcom": "ModelManager_ActiveBundle",
+  "usbgpu": "ModelManager_ActiveBundleUSBGPU",
+}
+_LAST_VALIDATED_RAW: dict[str, dict | None] = {}
 
 
 def _compute_hash(file_path: str) -> str | None:
@@ -86,11 +92,11 @@ def _bundle_needs_reset(active_bundle: custom.ModelManagerSP.ModelBundle, availa
   if available_bundles is not None:
     matching_bundle = None
     for bundle in available_bundles:
-      if getattr(active_bundle, 'ref', None) and getattr(bundle, 'ref', None):
+      if active_bundle.ref and bundle.ref:
         if active_bundle.ref == bundle.ref:
           matching_bundle = bundle
           break
-      elif getattr(active_bundle, 'internalName', None) == getattr(bundle, 'internalName', None):
+      elif active_bundle.internalName == bundle.internalName:
         matching_bundle = bundle
         break
 
@@ -98,49 +104,81 @@ def _bundle_needs_reset(active_bundle: custom.ModelManagerSP.ModelBundle, availa
       return True
     if active_bundle.minimumSelectorVersion != matching_bundle.minimumSelectorVersion:
       return True
-
-    active_runner = getattr(active_bundle, 'runner', None)
-    matching_runner = getattr(matching_bundle, 'runner', None)
-    if active_runner is not None and matching_runner is not None:
-      if getattr(active_runner, 'raw', active_runner) != getattr(matching_runner, 'raw', matching_runner):
-        return True
+    if active_bundle.runner != matching_bundle.runner:
+      return True
     if set(_bundle_artifacts(active_bundle)) != set(_bundle_artifacts(matching_bundle)):
       return True
 
   return not _bundle_is_valid_locally(active_bundle)
 
 
-def validate_active_bundle(params: Params, available_bundles: list[custom.ModelManagerSP.ModelBundle] | None = None) -> None:
-  global _LAST_VALIDATED_RAW
-
-  raw_bundle = params.get("ModelManager_ActiveBundle")
-  if not raw_bundle:
-    return
-
-  if raw_bundle == _LAST_VALIDATED_RAW:
-    return
-
-  active_bundle = get_active_bundle(params, raw_bundle_dict=raw_bundle)
-  if active_bundle is None or _bundle_needs_reset(active_bundle, available_bundles):
-    cloudlog.warning("Active model bundle invalid; resetting to default")
-    params.remove("ModelManager_ActiveBundle")
-    params.put("ModelRunnerTypeCache", int(custom.ModelManagerSP.Runner.stock), block=True)
-    _LAST_VALIDATED_RAW = None
-  else:
-    _LAST_VALIDATED_RAW = raw_bundle
-
-
-def get_active_bundle(params: Params | None = None, raw_bundle_dict: dict | bytes | None = None) -> "custom.ModelManagerSP.ModelBundle | None":
-  params = params or Params()
+def _parse_active_bundle(raw_bundle) -> "custom.ModelManagerSP.ModelBundle | None":
   try:
-    active_bundle_dict = raw_bundle_dict if raw_bundle_dict is not None else (params.get("ModelManager_ActiveBundle") or {})
-    if isinstance(active_bundle_dict, dict) and active_bundle_dict and is_bundle_version_compatible(active_bundle_dict):
-      return custom.ModelManagerSP.ModelBundle(**active_bundle_dict)
+    if isinstance(raw_bundle, dict) and raw_bundle and is_bundle_version_compatible(raw_bundle):
+      return custom.ModelManagerSP.ModelBundle(**raw_bundle)
   except Exception:
     pass
   return None
 
 
+def get_selected_bundle(params: Params | None = None, source: str = "qcom") -> "custom.ModelManagerSP.ModelBundle | None":
+  params = params or Params()
+  return _parse_active_bundle(params.get(ACTIVE_BUNDLE_KEYS[source]))
+
+
+def get_active_source(usbgpu: bool | None = None, usbgpu_active: bool | None = None,
+                      usbgpu_loading: bool | None = None, offroad: bool | None = None) -> str:
+  if usbgpu is None:
+    usbgpu = usbgpu_present()
+  state_valid = usbgpu_active is not None or usbgpu_loading is not None or offroad is not None
+  big_active = usbgpu and (not state_valid or usbgpu_active or usbgpu_loading or offroad)
+  return "usbgpu" if big_active else "qcom"
+
+
+def get_active_bundle(params: Params | None = None, *, usbgpu: bool | None = None) -> "custom.ModelManagerSP.ModelBundle | None":
+  # no cross-slot fallback: an empty active slot means the hardware default, which
+  # only stock modeld can run - modeld_v2 requires a real bundle
+  params = params or Params()
+  return get_selected_bundle(params, get_active_source(usbgpu=usbgpu))
+
+
+def resolve_bundle_by_ref(
+  ref: str, source_bundles: dict[str, list[custom.ModelManagerSP.ModelBundle]],
+) -> "tuple[custom.ModelManagerSP.ModelBundle, str] | None":
+  for source, bundles in source_bundles.items():
+    for bundle in bundles:
+      if bundle.ref == ref:
+        return bundle, source
+  return None
+
+
+def _validate_active_bundle(params: Params, source: str, available_bundles: list[custom.ModelManagerSP.ModelBundle] | None = None) -> None:
+  global _LAST_VALIDATED_RAW
+
+  key = ACTIVE_BUNDLE_KEYS[source]
+  raw_bundle = params.get(key)
+  if not raw_bundle:
+    return
+
+  if _LAST_VALIDATED_RAW.get(key) == raw_bundle:
+    return
+
+  active_bundle = _parse_active_bundle(raw_bundle)
+  if active_bundle is None or _bundle_needs_reset(active_bundle, available_bundles):
+    cloudlog.warning(f"Active model bundle invalid for {source}; resetting to default")
+    params.remove(key)
+    _LAST_VALIDATED_RAW[key] = None
+  else:
+    _LAST_VALIDATED_RAW[key] = raw_bundle
+
+
+def validate_active_bundles(params: Params, source_bundles: dict[str, list[custom.ModelManagerSP.ModelBundle]]) -> None:
+  # an empty list means the fetch failed, not that the catalog dropped the bundle
+  for source, bundles in source_bundles.items():
+    _validate_active_bundle(params, source, bundles or None)
+  get_active_model_runner(params, force_check=True)
+
+
 def get_active_model_runner(params: Params | None = None, force_check: bool = False) -> int:
   params = params or Params()
   cached_runner_type = params.get("ModelRunnerTypeCache")

openpilot/sunnypilot/models/manager.py

@@ -17,7 +17,8 @@ from openpilot.common.hardware.hw import Paths
 
 from openpilot.cereal import messaging, custom
 from openpilot.sunnypilot.models.fetcher import ModelFetcher
-from openpilot.sunnypilot.models.helpers import get_active_bundle, validate_active_bundle, verify_file
+from openpilot.sunnypilot.models.helpers import (ACTIVE_BUNDLE_KEYS, get_active_bundle, get_selected_bundle,
+                                                  resolve_bundle_by_ref, validate_active_bundles, verify_file)
 
 # (connect, read) seconds. read is per-request inactivity, not a total cap
 DOWNLOAD_TIMEOUT = (30, 30)
@@ -30,9 +31,12 @@ class ModelManagerSP:
     self.params = Params()
     self.model_fetcher = ModelFetcher(self.params)
     self.pm = messaging.PubMaster(["modelManagerSP"])
+    self.sm = messaging.SubMaster(["deviceState"])
+    self.chestnut_present = False
     self.available_models: list[custom.ModelManagerSP.ModelBundle] = []
+    self.source_models: dict[str, list[custom.ModelManagerSP.ModelBundle]] = {}
     self.selected_bundle: custom.ModelManagerSP.ModelBundle = None
-    self.active_bundle: custom.ModelManagerSP.ModelBundle = get_active_bundle(self.params)
+    self.active_bundle: custom.ModelManagerSP.ModelBundle = get_active_bundle(self.params, usbgpu=self.chestnut_present)
     self._chunk_size = 128 * 1000  # 128 KB chunks
     self._download_start_times: dict[str, float] = {}  # Track start time per model
 
@@ -76,7 +80,7 @@ class ModelManagerSP:
           f.write(chunk)
           bytes_downloaded += len(chunk)
 
-          if self.params.get("ModelManager_DownloadIndex") is None:
+          if self.params.get("ModelManager_DownloadRef") is None:
             raise Exception("Download cancelled")
 
           if total_size > 0:
@@ -114,7 +118,7 @@ class ModelManagerSP:
             for data in response.iter_content(chunk_size=self._chunk_size):
               f.write(data)
               chunk_downloaded += len(data)
-              if self.params.get("ModelManager_DownloadIndex") is None:
+              if self.params.get("ModelManager_DownloadRef") is None:
                 raise Exception("Download cancelled")
               intra = chunk_downloaded / max(chunk_size, 1)
               progress = min(99.0, ((i + intra) / num_chunks) * 100)
@@ -216,8 +220,7 @@ class ModelManagerSP:
     model_manager_state.availableBundles = self.available_models
     self.pm.send('modelManagerSP', msg)
 
-  async def _download_bundle(self, model_bundle: custom.ModelManagerSP.ModelBundle, destination_path: str) -> None:
-    """Downloads all models in a bundle"""
+  async def _download_bundle(self, model_bundle: custom.ModelManagerSP.ModelBundle, destination_path: str, source: str) -> None:
     self.selected_bundle = model_bundle
     self.selected_bundle.status = custom.ModelManagerSP.DownloadStatus.downloading
     for model in self.selected_bundle.models:
@@ -239,10 +242,9 @@ class ModelManagerSP:
           seen_artifacts.add(artifact.fileName)
           await self._process_artifact(artifact, destination_path)
 
-      self.active_bundle = self.selected_bundle
-      self.active_bundle.status = custom.ModelManagerSP.DownloadStatus.downloaded
-      self.params.put("ModelManager_ActiveBundle", self.active_bundle.to_dict(), block=True)
-      self.selected_bundle = None
+      self.selected_bundle.status = custom.ModelManagerSP.DownloadStatus.downloaded
+      self.params.put(ACTIVE_BUNDLE_KEYS[source], model_bundle.to_dict(), block=True)
+      self.active_bundle = get_active_bundle(self.params, usbgpu=self.chestnut_present)
 
     except Exception:
       if self.selected_bundle is not None:
@@ -252,9 +254,9 @@ class ModelManagerSP:
     finally:
       self._report_status()
 
-  def download(self, model_bundle: custom.ModelManagerSP.ModelBundle, destination_path: str) -> None:
+  def download(self, model_bundle: custom.ModelManagerSP.ModelBundle, destination_path: str, source: str) -> None:
     """Main entry point for downloading a model bundle"""
-    asyncio.run(self._download_bundle(model_bundle, destination_path))
+    asyncio.run(self._download_bundle(model_bundle, destination_path, source))
 
   def main_thread(self) -> None:
     """Main thread for model management"""
@@ -262,20 +264,22 @@ class ModelManagerSP:
 
     while True:
       try:
-        self.available_models = self.model_fetcher.get_available_bundles()
-        validate_active_bundle(self.params, self.available_models)
-        self.active_bundle = get_active_bundle(self.params)
+        self.sm.update(0)
+        self.chestnut_present = self.sm['deviceState'].chestnutPresent
+        self.source_models = {source: self.model_fetcher.get_bundles_for_source(source) for source in ModelFetcher.MODEL_SOURCES}
+        self.available_models = self.source_models[ModelFetcher.active_source(self.chestnut_present)]
+        validate_active_bundles(self.params, self.source_models)
+        self.active_bundle = get_active_bundle(self.params, usbgpu=self.chestnut_present)
 
-        if (index_to_download := self.params.get("ModelManager_DownloadIndex")) is not None:
-          if self.active_bundle and self.active_bundle.index == index_to_download:
-            self.params.remove("ModelManager_DownloadIndex")
-          elif model_to_download := next((model for model in self.available_models if model.index == index_to_download), None):
+        if (ref_to_download := self.params.get("ModelManager_DownloadRef")) is not None:
+          if resolved := resolve_bundle_by_ref(ref_to_download, self.source_models):
+            model_to_download, source = resolved
             try:
-              self.download(model_to_download, Paths.model_root())
+              self.download(model_to_download, Paths.model_root(), source)
             except Exception as e:
               cloudlog.exception(e)
             finally:
-              self.params.remove("ModelManager_DownloadIndex")
+              self.params.remove("ModelManager_DownloadRef")
               self.selected_bundle = None
 
         if self.params.get("ModelManager_ClearCache"):
@@ -294,12 +298,14 @@ class ModelManagerSP:
     Clears the model cache directory of all files except those in the active model bundle.
     """
 
-    # Get list of files used by active model bundle
+    # Get list of files used by both slots' selected bundles (either may become
+    # the truly active bundle depending on hardware availability)
     active_files = []
-    if self.active_bundle is not None: # When the default model is active
-      for model in self.active_bundle.models:
-        if hasattr(model, 'artifact') and model.artifact.fileName:
-          active_files.append(model.artifact.fileName)
+    for source in ACTIVE_BUNDLE_KEYS:
+      if selected_bundle := get_selected_bundle(self.params, source):
+        for model in selected_bundle.models:
+          if model.artifact.fileName:
+            active_files.append(model.artifact.fileName)
 
     # Remove all files except active ones (including their chunk files)
     model_dir = Paths.model_root()

openpilot/sunnypilot/models/tests/test_manager_download.py

@@ -11,6 +11,7 @@ import http.server
 import os
 import tempfile
 import threading
+import time
 import unittest
 from typing import Any
 from unittest import mock
@@ -23,6 +24,10 @@ from openpilot.common.test import OpenpilotTestCase
 from openpilot.common.file_chunker import get_chunk_name, get_manifest_path
 from openpilot.selfdrive.test.helpers import http_server_context
 from openpilot.sunnypilot.models import manager as manager_module
+from openpilot.sunnypilot.models.fetcher import ModelFetcher, get_cached_bundles
+from openpilot.sunnypilot.models import helpers
+from openpilot.sunnypilot.models.helpers import (get_active_bundle, get_active_source, get_selected_bundle,
+                                                  resolve_bundle_by_ref, validate_active_bundles)
 from openpilot.sunnypilot.models.manager import ModelManagerSP
 
 CHUNK_BODIES = [b'A' * 5000, b'B' * 5000, b'C' * 3000]
@@ -103,6 +108,7 @@ class ManagerDownloadTestBase(OpenpilotTestCase):
     self.manager.selected_bundle = None
     self.manager.active_bundle = None
     self.manager.available_models = []
+    self.manager.chestnut_present = False
     self.manager._chunk_size = 1024
     self.manager._download_start_times = {}
 
@@ -249,6 +255,85 @@ class TestManagerDownload(ManagerDownloadTestBase):
         assert self.manager._download_start_times == {}
     self.run_with_server(body)
 
+  def test_download_ref_present_keeps_download_alive(self):
+    """A pending download request (DownloadRef set) must not be cancelled mid-transfer."""
+    def body():
+      artifact = self.make_artifact(chunked=True)
+      base_path = os.path.join(self.dest, artifact.fileName)
+      self.manager.params.get.side_effect = lambda key: b"ref" if key == "ModelManager_DownloadRef" else None
+      asyncio.run(self.manager._download_chunked(artifact.downloadUri.uri, base_path, artifact))
+      assert os.path.isfile(get_manifest_path(base_path))
+    self.run_with_server(body)
+
+  def test_cancellation_via_download_ref(self):
+    """Removing DownloadRef mid-transfer cancels the download."""
+    def body():
+      artifact = self.make_artifact(chunked=True)
+      base_path = os.path.join(self.dest, artifact.fileName)
+      checks = {"n": 0}
+
+      def get(key):
+        if key == "ModelManager_DownloadRef":
+          checks["n"] += 1
+          return b"ref" if checks["n"] <= 2 else None
+        return b"0"
+
+      self.manager.params.get.side_effect = get
+      with self.assertRaises(Exception) as ctx:
+        asyncio.run(self.manager._download_chunked(artifact.downloadUri.uri, base_path, artifact))
+      assert 'cancelled' in str(ctx.exception).lower()
+      assert not os.path.isfile(get_manifest_path(base_path))
+    self.run_with_server(body)
+
+  def _make_params_with_store(self):
+    params = mock.MagicMock()
+    store = {}
+
+    def get(key, *args, **kwargs):
+      return store.get(key, b"0")  # b"0" -> download not cancelled
+
+    def put(key, value, *args, **kwargs):
+      store[key] = value
+
+    params.get.side_effect = get
+    params.put.side_effect = put
+    return params, store
+
+  def test_download_writes_qcom_slot(self):
+    """A download resolved to the qcom source writes the qcom active bundle slot only."""
+    def body():
+      artifact = self.make_artifact(chunked=True)
+      self._bundle.ref = "test-ref"
+      self._bundle.minimumSelectorVersion = 18
+      params, store = self._make_params_with_store()
+      self.manager.params = params
+      asyncio.run(self.manager._download_bundle(self._bundle, self.dest, "qcom"))
+
+      assert "ModelManager_ActiveBundle" in store, "qcom download must write the qcom slot"
+      assert "ModelManager_ActiveBundleUSBGPU" not in store, "qcom download must not touch the usbgpu slot"
+      assert self.manager.selected_bundle.status == custom.ModelManagerSP.DownloadStatus.downloaded
+      assert self.manager.active_bundle is not None and self.manager.active_bundle.ref == "test-ref"
+      assert self.manager.active_bundle.status == custom.ModelManagerSP.DownloadStatus.downloaded
+      chunk_names = [get_chunk_name(artifact.fileName, i, len(artifact.chunks)) for i in range(len(artifact.chunks))]
+      missing = [c for c in chunk_names if not os.path.isfile(os.path.join(self.dest, c))]
+      assert missing == [], f"chunks missing from the cache: {missing}"
+    self.run_with_server(body)
+
+  def test_download_writes_usbgpu_slot(self):
+    """A download resolved to the usbgpu source writes the usbgpu active bundle slot only."""
+    def body():
+      self.make_artifact(chunked=True)
+      self._bundle.ref = "big-ref"
+      self._bundle.minimumSelectorVersion = 18
+      params, store = self._make_params_with_store()
+      self.manager.params = params
+      asyncio.run(self.manager._download_bundle(self._bundle, self.dest, "usbgpu"))
+
+      assert "ModelManager_ActiveBundleUSBGPU" in store, "usbgpu download must write the usbgpu slot"
+      assert "ModelManager_ActiveBundle" not in store, "usbgpu download must not touch the qcom slot"
+      assert self.manager.selected_bundle.status == custom.ModelManagerSP.DownloadStatus.downloaded
+    self.run_with_server(body)
+
 
 class TestManagerImports(OpenpilotTestCase):
   """Catches undeclared dependencies. aiohttp lived only in the AGNOS venv; 19.6 dropped
@@ -267,6 +352,352 @@ class TestManagerImports(OpenpilotTestCase):
     assert connect > 0 and read > 0, "requests defaults to no timeout; downloads would hang forever"
 
 
+class TestResolveBundleByRef(OpenpilotTestCase):
+  """A ref resolves to (bundle, source) across both hardware manifests. Refs are
+  unique per manifest and never overlap across sources, so a ref maps to exactly
+  one slot. Shared by the manager's download flow and the settings UI."""
+
+  @staticmethod
+  def _bundle(ref: str):
+    bundle = custom.ModelManagerSP.ModelBundle.new_message()
+    bundle.ref = ref
+    return bundle
+
+  def test_qcom_ref_resolves_to_qcom_slot(self):
+    small = self._bundle("small")
+    assert resolve_bundle_by_ref("small", {"qcom": [small], "usbgpu": []}) == (small, "qcom")
+
+  def test_usbgpu_ref_resolves_to_usbgpu_slot(self):
+    big = self._bundle("big")
+    assert resolve_bundle_by_ref("big", {"qcom": [], "usbgpu": [big]}) == (big, "usbgpu")
+
+  def test_unknown_ref_returns_none(self):
+    source_bundles = {"qcom": [self._bundle("small")], "usbgpu": []}
+    assert resolve_bundle_by_ref("nope", source_bundles) is None
+
+
+def manifest_bundle(short_name: str, ref: str, index: int = 0, is_big: bool = False) -> dict:
+  """Minimal manifest bundle dict, version-compatible (no chunks to avoid disk side effects).
+  Big (usbgpu) bundles carry `is_big: true` in the manifest JSON."""
+  return {
+    "index": index,
+    "short_name": short_name,
+    "display_name": short_name.upper(),
+    "generation": 1,
+    "environment": "release",
+    "runner": "tinygrad",
+    "is_big": is_big,
+    "minimum_selector_version": "18",
+    "ref": ref,
+    "models": [{
+      "type": "supercombo",
+      "artifact": {
+        "file_name": f"{short_name}.pkl",
+        "download_uri": {"url": f"https://example.com/{short_name}.pkl", "sha256": "s"},
+      },
+    }],
+  }
+
+
+def fresh_sync_time() -> int:
+  return int(time.monotonic() * 1e9)
+
+
+class TestModelFetcherSources(OpenpilotTestCase):
+  """Both manifests are always maintained: get_bundles_for_source exposes either
+  source by name, and active_source picks which one matches the attached hardware."""
+
+  def _make_params(self, qcom_manifest, usbgpu_manifest):
+    params = mock.MagicMock()
+
+    def get(key):
+      if key == "ModelManager_ModelsCache":
+        return qcom_manifest
+      if key == "ModelManager_ModelsCache_USBGPU":
+        return usbgpu_manifest
+      if key in ("ModelManager_LastSyncTime", "ModelManager_LastSyncTime_USBGPU"):
+        return fresh_sync_time()
+      return None
+
+    params.get.side_effect = get
+    return params
+
+  def test_active_source_follows_chestnut_presence(self):
+    assert ModelFetcher.active_source(False) == "qcom"
+    assert ModelFetcher.active_source(True) == "usbgpu"
+
+  def test_get_bundles_for_source_returns_each_source(self):
+    params = self._make_params({"bundles": [manifest_bundle("small", "aaa")]},
+                               {"bundles": [manifest_bundle("big", "bbb", is_big=True)]})
+    fetcher = ModelFetcher(params)
+    assert [bundle.ref for bundle in fetcher.get_bundles_for_source("qcom")] == ["aaa"]
+    assert [bundle.ref for bundle in fetcher.get_bundles_for_source("usbgpu")] == ["bbb"]
+
+  def test_get_bundles_for_source_unknown(self):
+    assert ModelFetcher(mock.MagicMock()).get_bundles_for_source("bogus") == []
+
+  def test_get_cached_bundles_parses_source(self):
+    params = self._make_params({"bundles": [manifest_bundle("small", "aaa")]},
+                               {"bundles": [manifest_bundle("big", "bbb", is_big=True)]})
+    qcom_bundles = get_cached_bundles(params, "qcom")
+    usbgpu_bundles = get_cached_bundles(params, "usbgpu")
+    assert [b.ref for b in qcom_bundles] == ["aaa"]
+    assert [b.ref for b in usbgpu_bundles] == ["bbb"]
+    assert qcom_bundles[0].displayName == "SMALL"
+
+  def test_get_cached_bundles_empty_when_missing(self):
+    params = mock.MagicMock()
+    params.get.return_value = None
+    assert get_cached_bundles(params, "qcom") == []
+    assert get_cached_bundles(params, "usbgpu") == []
+
+  def test_get_cached_bundles_unknown_source(self):
+    assert get_cached_bundles(mock.MagicMock(), "bogus") == []
+
+  def test_active_json_has_both_urls(self):
+    params = mock.MagicMock()
+    ModelFetcher(params)
+    active_json_calls = [call for call in params.put.call_args_list if call.args[0] == "ModelManager_ActiveJson"]
+    assert active_json_calls, "expected ModelManager_ActiveJson to be written"
+    assert active_json_calls[-1].args[1] == {
+      "qcom": ModelFetcher.MODEL_URL,
+      "usbgpu": ModelFetcher.MODEL_URL_USBGPU,
+    }
+
+
+
+class TestSourceCacheIntegrity(OpenpilotTestCase):
+  """Each source's cached manifest must contain only that source's models; the
+  `is_big` flag in the JSON marks the big (usbgpu) models. A mismatched cache is
+  legacy data from before the per-source split (the active manifest was cached
+  under the unsuffixed key regardless of hardware) and is refetched. This
+  replaces the old one-time bundle migration."""
+
+  def _make_params(self, qcom_manifest, usbgpu_manifest):
+    params = mock.MagicMock()
+
+    def get(key):
+      if key == "ModelManager_ModelsCache":
+        return qcom_manifest
+      if key == "ModelManager_ModelsCache_USBGPU":
+        return usbgpu_manifest
+      if key in ("ModelManager_LastSyncTime", "ModelManager_LastSyncTime_USBGPU"):
+        return fresh_sync_time()
+      return None
+
+    params.get.side_effect = get
+    return params
+
+  def _fetched(self, *bundles):
+    return ModelFetcher(mock.MagicMock()).model_parser.parse_models({"bundles": list(bundles)})
+
+  def test_qcom_cache_with_big_models_is_refetched(self):
+    """Legacy: the unsuffixed cache holds the big manifest. is_big confirms it is
+    the wrong set for qcom, so a fresh fetch replaces it."""
+    params = self._make_params({"bundles": [manifest_bundle("big", "bbb", is_big=True)]},
+                               {"bundles": [manifest_bundle("big2", "ccc", is_big=True)]})
+    fetcher = ModelFetcher(params)
+    fetched = self._fetched(manifest_bundle("small", "aaa"))
+    with mock.patch.object(fetcher, "_fetch_and_cache_models", return_value=fetched):
+      bundles = fetcher.get_bundles_for_source("qcom")
+    assert [bundle.ref for bundle in bundles] == ["aaa"]
+
+  def test_usbgpu_cache_without_big_models_is_refetched(self):
+    params = self._make_params({"bundles": [manifest_bundle("small", "aaa")]},
+                               {"bundles": [manifest_bundle("big2", "ccc")]})
+    fetcher = ModelFetcher(params)
+    fetched = self._fetched(manifest_bundle("big", "bbb", is_big=True))
+    with mock.patch.object(fetcher, "_fetch_and_cache_models", return_value=fetched):
+      bundles = fetcher.get_bundles_for_source("usbgpu")
+    assert [bundle.ref for bundle in bundles] == ["bbb"]
+
+  def test_matching_caches_are_used_without_fetch(self):
+    params = self._make_params({"bundles": [manifest_bundle("small", "aaa")]},
+                               {"bundles": [manifest_bundle("big", "bbb", is_big=True)]})
+    fetcher = ModelFetcher(params)
+    with mock.patch.object(fetcher, "_fetch_and_cache_models", side_effect=AssertionError("cache should be used")):
+      assert [bundle.ref for bundle in fetcher.get_bundles_for_source("qcom")] == ["aaa"]
+      assert [bundle.ref for bundle in fetcher.get_bundles_for_source("usbgpu")] == ["bbb"]
+
+  def test_stale_version_cache_is_refetched(self):
+    """A source-matching cache whose bundles are all filtered by the selector version
+    check parses to zero valid bundles; it is stale (e.g. an old manifest) and must be
+    refetched instead of silently returning an empty list forever."""
+    stale = manifest_bundle("small", "aaa")
+    stale["minimum_selector_version"] = "16"
+    params = self._make_params({"bundles": [stale]},
+                               {"bundles": [manifest_bundle("big", "bbb", is_big=True)]})
+    fetcher = ModelFetcher(params)
+    fetched = self._fetched(manifest_bundle("small2", "ddd"))
+    with mock.patch.object(fetcher, "_fetch_and_cache_models", return_value=fetched) as fetch:
+      bundles = fetcher.get_bundles_for_source("qcom")
+    fetch.assert_called_once_with("qcom")
+    assert [bundle.ref for bundle in bundles] == ["ddd"]
+
+  def test_mismatched_refetch_happens_once(self):
+    """If the fresh manifest still fails the source check, the URL is authoritative:
+    trust it instead of refetching at 1 Hz forever."""
+    params = self._make_params({"bundles": [manifest_bundle("big", "bbb", is_big=True)]},
+                               {"bundles": [manifest_bundle("big2", "ccc", is_big=True)]})
+    fetcher = ModelFetcher(params)
+    fetched = self._fetched(manifest_bundle("big", "bbb", is_big=True))
+    with mock.patch.object(fetcher, "_fetch_and_cache_models", return_value=fetched) as fetch:
+      first = fetcher.get_bundles_for_source("qcom")
+      second = fetcher.get_bundles_for_source("qcom")
+    fetch.assert_called_once_with("qcom")
+    assert [bundle.ref for bundle in first] == ["bbb"]
+    assert [bundle.ref for bundle in second] == ["bbb"]
+
+  def test_corrupt_cache_is_refetched(self):
+    """A cache that fails to parse (e.g. truncated/foreign JSON) must trigger a
+    refetch instead of raising every loop and never recovering."""
+    corrupt = {"bundles": [{"short_name": "broken"}]}  # missing required fields
+    params = self._make_params(corrupt, {"bundles": [manifest_bundle("big", "bbb", is_big=True)]})
+    fetcher = ModelFetcher(params)
+    fetched = self._fetched(manifest_bundle("small", "aaa"))
+    with mock.patch.object(fetcher, "_fetch_and_cache_models", return_value=fetched) as fetch:
+      bundles = fetcher.get_bundles_for_source("qcom")
+    fetch.assert_called_once_with("qcom")
+    assert [bundle.ref for bundle in bundles] == ["aaa"]
+
+
+class TestActiveBundleValidation(OpenpilotTestCase):
+  """Validation is per-slot: a failed fetch (empty bundle list) must not reset a slot,
+  and resetting one slot must not stomp the runner cache derived from the other."""
+
+  def setUp(self):
+    super().setUp()
+    helpers._LAST_VALIDATED_RAW.clear()
+
+  @staticmethod
+  def _raw_bundle(ref: str, runner: int | None = None) -> dict:
+    bundle = custom.ModelManagerSP.ModelBundle.new_message()
+    bundle.ref = ref
+    bundle.minimumSelectorVersion = 18
+    if runner is not None:
+      bundle.runner = runner
+    return bundle.to_dict()
+
+  def _params(self, qcom=None, usbgpu=None):
+    params = mock.MagicMock()
+
+    def get(key, *args, **kwargs):
+      return {"ModelManager_ActiveBundle": qcom, "ModelManager_ActiveBundleUSBGPU": usbgpu}.get(key)
+
+    params.get.side_effect = get
+    return params
+
+  def test_empty_catalog_does_not_reset_slot(self):
+    params = self._params(qcom=self._raw_bundle("small"))
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=False):
+      validate_active_bundles(params, {"qcom": [], "usbgpu": []})
+    params.remove.assert_not_called()
+
+  def test_reset_recomputes_runner_from_surviving_slot(self):
+    tinygrad = int(custom.ModelManagerSP.Runner.tinygrad)
+    big_raw = self._raw_bundle("big", runner=tinygrad)
+    params = self._params(qcom=self._raw_bundle("gone"), usbgpu=big_raw)
+    catalog = {"qcom": [custom.ModelManagerSP.ModelBundle(**self._raw_bundle("other"))],
+               "usbgpu": [custom.ModelManagerSP.ModelBundle(**big_raw)]}
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=True):
+      validate_active_bundles(params, catalog)
+    params.remove.assert_called_once_with("ModelManager_ActiveBundle")
+    runner_puts = [call for call in params.put.call_args_list if call.args[0] == "ModelRunnerTypeCache"]
+    assert [call.args[1] for call in runner_puts] == [tinygrad]
+
+
+class TestActiveBundleSelection(OpenpilotTestCase):
+  """The effective active bundle is the active source's slot: usbgpu when a GPU is
+  present, qcom otherwise. An empty active slot means the hardware default (stock
+  runner), never the other slot's pick - modeld_v2 requires a real bundle."""
+
+  @staticmethod
+  def _raw_bundle(ref: str) -> dict:
+    bundle = custom.ModelManagerSP.ModelBundle.new_message()
+    bundle.ref = ref
+    bundle.minimumSelectorVersion = 18
+    return bundle.to_dict()
+
+  def _params(self, qcom=None, usbgpu=None):
+    params = mock.MagicMock()
+
+    def get(key, *args, **kwargs):
+      if key == "ModelManager_ActiveBundle":
+        return qcom
+      if key == "ModelManager_ActiveBundleUSBGPU":
+        return usbgpu
+      return None
+
+    params.get.side_effect = get
+    return params
+
+  def test_selected_bundle_is_per_slot(self):
+    params = self._params(qcom=self._raw_bundle("small"), usbgpu=self._raw_bundle("big"))
+    assert get_selected_bundle(params, "qcom").ref == "small"
+    assert get_selected_bundle(params, "usbgpu").ref == "big"
+
+  def test_no_gpu_uses_qcom_slot(self):
+    params = self._params(qcom=self._raw_bundle("small"), usbgpu=self._raw_bundle("big"))
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=False):
+      assert get_active_bundle(params).ref == "small"
+
+  def test_gpu_uses_usbgpu_slot(self):
+    params = self._params(qcom=self._raw_bundle("small"), usbgpu=self._raw_bundle("big"))
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=True):
+      assert get_active_bundle(params).ref == "big"
+
+  def test_gpu_without_big_selection_is_hardware_default(self):
+    params = self._params(qcom=self._raw_bundle("small"), usbgpu=None)
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=True):
+      assert get_active_bundle(params) is None
+
+
+class TestEffectiveSource(OpenpilotTestCase):
+  """One gate decides the active source. With no flags it is runtime truth (GPU
+  attached); display callers (mici) pass the ui_state flags, which additionally
+  require the big model to be loading, active, or the device offroad. The active
+  bundle is simply the selected bundle of that source."""
+
+  @staticmethod
+  def _raw_bundle(ref: str) -> dict:
+    bundle = custom.ModelManagerSP.ModelBundle.new_message()
+    bundle.ref = ref
+    bundle.minimumSelectorVersion = 18
+    return bundle.to_dict()
+
+  def test_runtime_no_gpu(self):
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=False):
+      assert get_active_source() == "qcom"
+
+  def test_runtime_gpu_present(self):
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=True):
+      assert get_active_source() == "usbgpu"
+
+  def test_display_offroad_gpu_present_shows_big(self):
+    assert get_active_source(usbgpu=True, usbgpu_active=False, usbgpu_loading=False, offroad=True) == "usbgpu"
+
+  def test_display_onroad_gpu_loading_shows_big(self):
+    assert get_active_source(usbgpu=True, usbgpu_active=False, usbgpu_loading=True, offroad=False) == "usbgpu"
+
+  def test_display_onroad_gpu_active_shows_big(self):
+    assert get_active_source(usbgpu=True, usbgpu_active=True, usbgpu_loading=False, offroad=False) == "usbgpu"
+
+  def test_display_onroad_gpu_idle_shows_small(self):
+    assert get_active_source(usbgpu=True, usbgpu_active=False, usbgpu_loading=False, offroad=False) == "qcom"
+
+  def test_display_active_none_is_idle(self):
+    assert get_active_source(usbgpu=True, usbgpu_active=None, usbgpu_loading=False, offroad=False) == "qcom"
+
+  def test_active_bundle_follows_source(self):
+    params = mock.MagicMock()
+    params.get.side_effect = lambda key: {"ModelManager_ActiveBundle": self._raw_bundle("small"),
+                                          "ModelManager_ActiveBundleUSBGPU": self._raw_bundle("big")}.get(key)
+    with mock.patch("openpilot.sunnypilot.models.helpers.usbgpu_present", return_value=False):
+      assert get_active_bundle(params).ref == "small"
+    assert get_selected_bundle(params, get_active_source(usbgpu=True, usbgpu_active=False,
+                                                         usbgpu_loading=False, offroad=True)).ref == "big"
+
+
 @unittest.skipUnless(os.environ.get('RUN_INTEGRATION_TESTS'), 'requires external network')
 class TestLiveModelManifest(OpenpilotTestCase):
   """Every artifact and chunk URL in the published manifest must resolve."""

openpilot/sunnypilot/models/tests/test_tinygrad_ref.py

@@ -1,13 +1,11 @@
 import requests
 
-from openpilot.common.params import Params
 from openpilot.sunnypilot.models.tinygrad_ref import get_tinygrad_ref
 from openpilot.sunnypilot.models.fetcher import ModelFetcher
 from openpilot.common.test import OpenpilotTestCase
 
 def fetch_tinygrad_ref():
-  fetcher = ModelFetcher(Params())
-  response = requests.get(fetcher.model_url, timeout=10)
+  response = requests.get(ModelFetcher.MODEL_URL, timeout=10)
   response.raise_for_status()
   json_data = response.json()
   return json_data.get("tinygrad_ref")

openpilot/sunnypilot/sunnylink/athena/sunnylinkd.py

@@ -28,7 +28,6 @@ from websocket import (ABNF, WebSocket, WebSocketException, WebSocketTimeoutExce
                        create_connection, WebSocketConnectionClosedException)
 
 import openpilot.cereal.messaging as messaging
-from openpilot.selfdrive.modeld.helpers import usbgpu_present
 from openpilot.sunnypilot.models.model_name import DEFAULT_MODEL, DEFAULT_BIG_MODEL
 from openpilot.sunnypilot.selfdrive.car.sync_sunnylink_params import update_car_list_param
 from openpilot.sunnypilot.sunnylink.api import SunnylinkApi
@@ -182,10 +181,8 @@ def getParamsMetadata() -> str:
     schema = generate_schema()
     schema["capabilities"] = generate_capabilities()
     schema["capability_labels"] = CAPABILITY_LABELS
-    # mirrors get_default_model() — ui_state unavailable in sunnylinkd process
-    show_big = (usbgpu_present()
-                and (params.get_bool("UsbGpuActive") or params.get_bool("UsbGpuLoading") or params.get_bool("IsOffroad")))
-    schema["default_model"] = DEFAULT_BIG_MODEL if show_big else DEFAULT_MODEL
+    schema["default_model"] = DEFAULT_MODEL
+    schema["default_big_model"] = DEFAULT_BIG_MODEL
     schema["usbgpu_active"] = params.get_bool("UsbGpuActive")
     raw = json.dumps(schema, separators=(",", ":")).encode("utf-8")
     return base64.b64encode(gzip.compress(raw)).decode("utf-8")

openpilot/sunnypilot/sunnylink/statsd.py

@@ -65,6 +65,7 @@ def sp_stats(end_event):
     'MadsSteeringMode',
     'MadsUnifiedEngagementMode',
     'ModelManager_ActiveBundle',
+    'ModelManager_ActiveBundleUSBGPU',
     'ModelManager_Favs',
     'EnableSunnylinkUploader',
     'SunnylinkEnabled',

openpilot/sunnypilot/system/params_migration.py

@@ -84,6 +84,21 @@ def _migrate_tesla_mads_screen_button(_params):
     cloudlog.exception(f"Error migrating TeslaMadsScreenButton: {e}")
 
 
+def _migrate_model_bundle_slots(_params):
+  # Pre-split, a chestnut user's big-model selection lived in the single
+  # ActiveBundle. Seed both slots; validation drops whichever does not match
+  # its own manifest.
+  try:
+    if _params.get("ModelManager_ActiveBundleUSBGPU") is not None:
+      return
+    if (bundle := _params.get("ModelManager_ActiveBundle")) is None:
+      return
+    _params.put("ModelManager_ActiveBundleUSBGPU", bundle, block=True)
+    cloudlog.info("params_migration: seeded ModelManager_ActiveBundleUSBGPU from ModelManager_ActiveBundle")
+  except Exception as e:
+    cloudlog.exception(f"Error migrating model bundle slots: {e}")
+
+
 def run_migration(_params):
   # migrate OnroadScreenOffBrightness
   if _params.get("OnroadScreenOffBrightnessMigrated") != ONROAD_BRIGHTNESS_MIGRATION_VERSION:
@@ -120,3 +135,6 @@ def run_migration(_params):
 
   # seed TeslaMadsScreenButton for existing Tesla installs
   _migrate_tesla_mads_screen_button(_params)
+
+  # seed the usbgpu model slot from the pre-split single slot
+  _migrate_model_bundle_slots(_params)

openpilot/sunnypilot/system/tests/test_params_migration.py

@@ -0,0 +1,36 @@
+"""
+Copyright (c) 2021-, Haibin Wen, sunnypilot, and a number of other contributors.
+
+This file is part of sunnypilot and is licensed under the MIT License.
+See the LICENSE.md file in the root directory for more details.
+"""
+
+from openpilot.common.params import Params
+from openpilot.common.test import OpenpilotTestCase
+from openpilot.sunnypilot.system.params_migration import _migrate_model_bundle_slots
+
+
+class TestModelBundleSlotMigration(OpenpilotTestCase):
+  """Pre-split, a chestnut user's big-model selection lived in the single ActiveBundle.
+  The migration seeds both slots; per-source validation later drops whichever does not
+  match its own manifest."""
+
+  def test_seeds_usbgpu_slot_from_active_bundle(self):
+    params = Params()
+    bundle = {"ref": "big", "minimumSelectorVersion": 18}
+    params.put("ModelManager_ActiveBundle", bundle, block=True)
+    _migrate_model_bundle_slots(params)
+    assert params.get("ModelManager_ActiveBundleUSBGPU") == bundle
+    assert params.get("ModelManager_ActiveBundle") == bundle
+
+  def test_noop_when_usbgpu_slot_already_set(self):
+    params = Params()
+    params.put("ModelManager_ActiveBundle", {"ref": "small"}, block=True)
+    params.put("ModelManager_ActiveBundleUSBGPU", {"ref": "big"}, block=True)
+    _migrate_model_bundle_slots(params)
+    assert params.get("ModelManager_ActiveBundleUSBGPU") == {"ref": "big"}
+
+  def test_noop_when_no_selection(self):
+    params = Params()
+    _migrate_model_bundle_slots(params)
+    assert params.get("ModelManager_ActiveBundleUSBGPU") is None

openpilot/system/ui/sunnypilot/lib/utils.py

@@ -8,12 +8,26 @@ from collections.abc import Callable
 
 import pyray as rl
 
-from openpilot.system.ui.lib.application import FontWeight
+from openpilot.system.ui.lib.application import gui_app, FontWeight
 from openpilot.system.ui.sunnypilot.lib.styles import style
 from openpilot.system.ui.sunnypilot.widgets.list_view import ButtonActionSP
-from openpilot.system.ui.widgets.label import UnifiedLabel
+from openpilot.system.ui.widgets.label import ScrollState, UnifiedLabel
 from openpilot.system.ui.widgets.list_view import BUTTON_WIDTH, BUTTON_HEIGHT, TEXT_PADDING, _resolve_value
 
+SCROLL_SPEED = 1.2  # stock is 0.8, boosted 50% to compensate for larger font (50 vs 32)
+SCROLL_REFERENCE_FPS = 60.
+
+
+class UnifiedLabelSP(UnifiedLabel):
+  # stock scroll formula (0.8 / 60 * fps) is inverted — pre-correct so speed is constant px/sec
+  def _render(self, _):
+    if self._needs_scroll and self._scroll_state == ScrollState.SCROLLING:
+      fps = gui_app.target_fps
+      wrong_step = 0.8 / SCROLL_REFERENCE_FPS * fps
+      correct_step = SCROLL_SPEED * SCROLL_REFERENCE_FPS / fps
+      self._scroll_offset -= (correct_step - wrong_step)
+    super()._render(_)
+
 
 class NoElideButtonAction(ButtonActionSP):
   def get_width_hint(self):
@@ -21,14 +35,12 @@ class NoElideButtonAction(ButtonActionSP):
 
 
 class ScrollingButtonAction(ButtonActionSP):
-  """ButtonActionSP whose value scrolls instead of eliding when it doesn't fit."""
-
   def __init__(self, text: str | Callable[[], str], width: int = style.BUTTON_ACTION_WIDTH,
                enabled: bool | Callable[[], bool] = True):
     super().__init__(text=text, width=width, enabled=enabled)
-    self._value_label = UnifiedLabel("", font_size=style.ITEM_TEXT_FONT_SIZE, font_weight=FontWeight.NORMAL,
-                                     text_color=self._value_color, scroll=True,
-                                     alignment_vertical=rl.GuiTextAlignmentVertical.TEXT_ALIGN_MIDDLE)
+    self._value_label = UnifiedLabelSP("", font_size=style.ITEM_TEXT_FONT_SIZE, font_weight=FontWeight.NORMAL,
+                                       text_color=self._value_color, scroll=True,
+                                       alignment_vertical=rl.GuiTextAlignmentVertical.TEXT_ALIGN_MIDDLE)
 
   def set_value(self, value: str | Callable[[], str], color: rl.Color = style.ITEM_TEXT_VALUE_COLOR):
     if self.value != _resolve_value(value, ""):

release/ci/model_generator.py

@@ -136,7 +136,7 @@ def generate_chunked_model(driving_pkl: Path) -> dict:
 
 
 def create_metadata_json(models: list, output_dir: Path, custom_name=None, short_name=None, is_20hz=False, upstream_branch="unknown",
-                         onnx_sha256=None) -> None:
+                         onnx_sha256=None, is_big=False) -> None:
   bundle_json = {
     "short_name": short_name,
     "display_name": custom_name or upstream_branch,
@@ -149,6 +149,7 @@ def create_metadata_json(models: list, output_dir: Path, custom_name=None, short
     "generation": "-1",
     "build_time": datetime.now(UTC).strftime("%Y-%m-%dT%H:%M:%SZ"),
     "overrides": {},
+    "is_big": is_big,
     "models": models,
   }
 
@@ -186,6 +187,8 @@ if __name__ == "__main__":
     print(f"No driving_tinygrad.pkl found in {_output_dir}", file=sys.stderr)
     sys.exit(1)
 
+  is_big = _driving_pkl.name.startswith('big_')
+
   if _pkl:
     new_pkl = _output_dir / f"driving_{_pkl}_tinygrad.pkl"
     if not new_pkl.exists():
@@ -196,4 +199,4 @@ if __name__ == "__main__":
   _model_metadata = generate_chunked_model(_driving_pkl)
   _onnx_sha256 = _hash_onnx_files(Path(args.model_dir))
   create_metadata_json([_model_metadata], _output_dir, args.custom_name, _short_name, args.is_20hz, args.upstream_branch,
-                       onnx_sha256=_onnx_sha256)
+                       onnx_sha256=_onnx_sha256, is_big=is_big)

release/ci/publish.sh

@@ -47,7 +47,7 @@ git rm -rf $OUTPUT_DIR/.git || true # Doing cleanup, but it might fail if the .g
 git remote remove origin || true # ensure cleanup
 git remote add origin $GIT_ORIGIN
 #git push origin -d $DEV_BRANCH || true # Ensuring we delete the remote branch if it exists as we are wiping it out
-git fetch origin $DEV_BRANCH || (git checkout -b $DEV_BRANCH && git commit --allow-empty -m "sunnypilot v$VERSION release" && git push -u origin $DEV_BRANCH)
+git fetch --depth 1 origin $DEV_BRANCH || (git checkout -b $DEV_BRANCH && git commit --allow-empty -m "sunnypilot v$VERSION release" && git push -u origin $DEV_BRANCH)
 
 echo "[-] committing version $VERSION T=$SECONDS"
 git add -f .