Move nested payloads to installer and zip file on windows

Now that the llm runner is an executable and not just a dll, more users are facing
problems with security policy configurations on windows that prevent users
writing to directories and then executing binaries from the same location.
This change removes payloads from the main executable on windows and shifts them
over to be packaged in the installer and discovered based on the executables location.
This also adds a new zip file for people who want to "roll their own" installation model.

.github/workflows/release.yaml

@@ -103,6 +103,7 @@ jobs:
           path: |
             llm/build/**/bin/*
             llm/build/**/*.a
+            dist/windows-amd64/**
 
   # ROCm generation step
   generate-windows-rocm:
@@ -173,7 +174,9 @@ jobs:
       - uses: actions/upload-artifact@v4
         with:
           name: generate-windows-rocm
-          path: llm/build/**/bin/*
+          path: |
+            llm/build/**/bin/*
+            dist/windows-amd64/**
       - uses: actions/upload-artifact@v4
         with:
           name: windows-rocm-deps
@@ -253,7 +256,9 @@ jobs:
       - uses: actions/upload-artifact@v4
         with:
           name: generate-windows-cuda
-          path: llm/build/**/bin/*
+          path: |
+            llm/build/**/bin/*
+            dist/windows-amd64/**
       - uses: actions/upload-artifact@v4
         with:
           name: windows-cuda-deps
@@ -306,11 +311,15 @@ jobs:
       - uses: actions/download-artifact@v4
         with:
           name: generate-windows-cpu
-          path: llm/build
+          path: |
+            llm/build
+            dist/windows-amd64
       - uses: actions/download-artifact@v4
         with:
           name: generate-windows-cuda
-          path: llm/build
+          path: |
+            llm/build
+            dist/windows-amd64
       - uses: actions/download-artifact@v4
         with:
           name: windows-cuda-deps
@@ -322,7 +331,9 @@ jobs:
       - uses: actions/download-artifact@v4
         with:
           name: generate-windows-rocm
-          path: llm/build
+          path: |
+            llm/build
+            dist/windows-amd64
       - run: dir llm/build
       - run: |
           $gopath=(get-command go).source | split-path -parent
@@ -337,7 +348,9 @@ jobs:
       - uses: actions/upload-artifact@v4
         with:
           name: dist-windows
-          path: dist/*.exe
+          path: |
+            dist/OllamaSetup.exe
+            dist/ollama-windows-*.zip
 
   # Linux x86 assets built using the container based build
   build-linux-amd64:

.github/workflows/test.yaml

@@ -103,7 +103,9 @@ jobs:
       - uses: actions/upload-artifact@v4
         with:
           name: cuda-${{ matrix.cuda-version }}-libraries
-          path: llm/build/**/bin/*
+          path: |
+            llm/build/**/bin/*
+            dist/windows-amd64/**
   generate-rocm:
     needs: [changes]
     if: ${{ needs.changes.outputs.GENERATE_ROCM == 'True' }}
@@ -134,7 +136,9 @@ jobs:
       - uses: actions/upload-artifact@v4
         with:
           name: rocm-${{ matrix.rocm-version }}-libraries
-          path: llm/build/**/bin/*
+          path: |
+            llm/build/**/bin/*
+            dist/windows-amd64/**
 
   # ROCm generation step
   generate-windows-rocm:
@@ -253,11 +257,6 @@ jobs:
           mkdir -p llm/build/darwin/$ARCH/stub/bin
           touch llm/build/darwin/$ARCH/stub/bin/ollama_llama_server
         if: ${{ startsWith(matrix.os, 'macos-') }}
-      - run: |
-          mkdir -p llm/build/windows/$ARCH/stub/bin
-          touch llm/build/windows/$ARCH/stub/bin/ollama_llama_server
-        if: ${{ startsWith(matrix.os, 'windows-') }}
-        shell: bash
       - uses: golangci/golangci-lint-action@v4
         with:
           args: --timeout 8m0s -v
@@ -299,10 +298,6 @@ jobs:
           mkdir -p llm/build/darwin/$ARCH/stub/bin
           touch llm/build/darwin/$ARCH/stub/bin/ollama_llama_server
         if: ${{ startsWith(matrix.os, 'macos-') }}
-      - run: |
-          mkdir -p llm/build/windows/$ARCH/stub/bin
-          touch llm/build/windows/$ARCH/stub/bin/ollama_llama_server
-        if: ${{ startsWith(matrix.os, 'windows-') }}
         shell: bash
       - run: go generate ./...
       - run: go build