Merge branch 'dev' into local-server-documentation

.github/scripts/auto-sign.sh

@@ -8,3 +8,5 @@ fi
 
 # If both variables are set, execute the following commands
 find "$APP_PATH" \( -type f -perm +111 -o -name "*.node" \) -exec codesign -s "$DEVELOPER_ID" --options=runtime {} \;
+
+find "$APP_PATH" -type f -name "*.o" -exec codesign -s "$DEVELOPER_ID" --options=runtime {} \;

.github/workflows/jan-electron-build-nightly.yml

@@ -48,8 +48,17 @@ jobs:
   get-update-version:
     uses: ./.github/workflows/template-get-update-version.yml
 
-  build-macos:
+  build-macos-x64:
-    uses: ./.github/workflows/template-build-macos.yml
+    uses: ./.github/workflows/template-build-macos-x64.yml
+    needs: [get-update-version, set-public-provider]
+    secrets: inherit
+    with:
+      ref: ${{ needs.set-public-provider.outputs.ref }}
+      public_provider: ${{ needs.set-public-provider.outputs.public_provider }}
+      new_version: ${{ needs.get-update-version.outputs.new_version }}
+
+  build-macos-arm64:
+    uses: ./.github/workflows/template-build-macos-arm64.yml
     needs: [get-update-version, set-public-provider]
     secrets: inherit
     with:
@@ -76,8 +85,51 @@ jobs:
       public_provider: ${{ needs.set-public-provider.outputs.public_provider }}
       new_version: ${{ needs.get-update-version.outputs.new_version }}
 
+  combine-latest-mac-yml:
+    needs: [set-public-provider, build-macos-x64, build-macos-arm64]
+    runs-on: ubuntu-latest
+    steps:
+      - name: Getting the repo
+        uses: actions/checkout@v3
+        with:
+          ref: ${{ needs.set-public-provider.outputs.ref }}
+      - name: Download mac-x64 artifacts
+        uses: actions/download-artifact@v3
+        with:
+          name: latest-mac-x64
+          path: ./latest-mac-x64
+      - name: Download mac-arm artifacts
+        uses: actions/download-artifact@v3
+        with:
+          name: latest-mac-arm64
+          path: ./latest-mac-arm64
+
+      - name: 'Merge latest-mac.yml'
+        # unfortunately electron-builder doesn't understand that we have two different releases for mac-x64 and mac-arm, so we need to manually merge the latest files
+        # see https://github.com/electron-userland/electron-builder/issues/5592
+        run: |
+          ls -la .
+          ls -la ./latest-mac-x64
+          ls -la ./latest-mac-arm64
+          ls -la ./electron
+          cp ./electron/merge-latest-ymls.js /tmp/merge-latest-ymls.js
+          npm install js-yaml --prefix /tmp
+          node /tmp/merge-latest-ymls.js ./latest-mac-x64/latest-mac.yml ./latest-mac-arm64/latest-mac.yml ./latest-mac.yml
+          cat ./latest-mac.yml
+      
+      - name: Upload latest-mac.yml
+        if: ${{ needs.set-public-provider.outputs.public_provider == 'cloudflare-r2' }}
+        run: |
+          aws s3api put-object --endpoint-url https://${{ secrets.CLOUDFLARE_ACCOUNT_ID }}.r2.cloudflarestorage.com --bucket ${{ secrets.CLOUDFLARE_R2_BUCKET_NAME }} --key "latest/latest-mac.yml" --body "./latest-mac.yml"
+        env:
+          AWS_ACCESS_KEY_ID: ${{ secrets.CLOUDFLARE_R2_ACCESS_KEY_ID }}
+          AWS_SECRET_ACCESS_KEY: ${{ secrets.CLOUDFLARE_R2_SECRET_ACCESS_KEY }}
+          AWS_DEFAULT_REGION: auto
+          AWS_EC2_METADATA_DISABLED: "true"
+
+    
   noti-discord-nightly-and-update-url-readme:
-    needs: [build-macos, build-windows-x64, build-linux-x64, get-update-version, set-public-provider]
+    needs: [build-macos-x64, build-macos-arm64, build-windows-x64, build-linux-x64, get-update-version, set-public-provider, combine-latest-mac-yml]
     secrets: inherit
     if: github.event_name == 'schedule'
     uses: ./.github/workflows/template-noti-discord-and-update-url-readme.yml
@@ -88,7 +140,7 @@ jobs:
       new_version: ${{ needs.get-update-version.outputs.new_version }}
 
   noti-discord-pre-release-and-update-url-readme:
-    needs: [build-macos, build-windows-x64, build-linux-x64, get-update-version, set-public-provider]
+    needs: [build-macos-x64, build-macos-arm64, build-windows-x64, build-linux-x64, get-update-version, set-public-provider, combine-latest-mac-yml]
     secrets: inherit
     if: github.event_name == 'push'
     uses: ./.github/workflows/template-noti-discord-and-update-url-readme.yml
@@ -99,7 +151,7 @@ jobs:
       new_version: ${{ needs.get-update-version.outputs.new_version }}
 
   noti-discord-manual-and-update-url-readme:
-    needs: [build-macos, build-windows-x64, build-linux-x64, get-update-version, set-public-provider]
+    needs: [build-macos-x64, build-macos-arm64, build-windows-x64, build-linux-x64, get-update-version, set-public-provider, combine-latest-mac-yml]
     secrets: inherit
     if: github.event_name == 'workflow_dispatch' && github.event.inputs.public_provider == 'cloudflare-r2'
     uses: ./.github/workflows/template-noti-discord-and-update-url-readme.yml

.github/workflows/jan-electron-build.yml

@@ -9,8 +9,42 @@ jobs:
   get-update-version:
     uses: ./.github/workflows/template-get-update-version.yml
 
-  build-macos:
+  create-draft-release:
-    uses: ./.github/workflows/template-build-macos.yml
+    runs-on: ubuntu-latest
+    if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/')
+    outputs:
+      upload_url: ${{ steps.create_release.outputs.upload_url }}
+      version: ${{ steps.get_version.outputs.version }}
+    permissions:
+      contents: write
+    steps:
+      - name: Extract tag name without v prefix
+        id: get_version
+        run: echo "VERSION=${GITHUB_REF#refs/tags/v}" >> $GITHUB_ENV && echo "::set-output name=version::${GITHUB_REF#refs/tags/v}"
+        env:
+          GITHUB_REF: ${{ github.ref }}
+      - name: Create Draft Release
+        id: create_release
+        uses: actions/create-release@v1
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        with:
+          tag_name: ${{ github.ref_name }}
+          release_name: "${{ env.VERSION }}"
+          draft: true
+          prerelease: false
+
+  build-macos-x64:
+    uses: ./.github/workflows/template-build-macos-x64.yml
+    secrets: inherit
+    needs: [get-update-version]
+    with:
+      ref: ${{ github.ref }}
+      public_provider: github
+      new_version: ${{ needs.get-update-version.outputs.new_version }}
+
+  build-macos-arm64:
+    uses: ./.github/workflows/template-build-macos-arm64.yml
     secrets: inherit
     needs: [get-update-version]
     with:
@@ -36,8 +70,52 @@ jobs:
       public_provider: github
       new_version: ${{ needs.get-update-version.outputs.new_version }}
 
+  combine-latest-mac-yml:
+    needs: [build-macos-x64, build-macos-arm64, create-draft-release]
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+    steps:
+      - name: Getting the repo
+        uses: actions/checkout@v3
+
+      - name: Download mac-x64 artifacts
+        uses: actions/download-artifact@v3
+        with:
+          name: latest-mac-x64
+          path: ./latest-mac-x64
+      - name: Download mac-arm artifacts
+        uses: actions/download-artifact@v3
+        with:
+          name: latest-mac-arm64
+          path: ./latest-mac-arm64
+
+      - name: 'Merge latest-mac.yml'
+        # unfortunately electron-builder doesn't understand that we have two different releases for mac-x64 and mac-arm, so we need to manually merge the latest files
+        # see https://github.com/electron-userland/electron-builder/issues/5592
+        run: |
+          ls -la .
+          ls -la ./latest-mac-x64
+          ls -la ./latest-mac-arm64
+          ls -la ./electron
+          cp ./electron/merge-latest-ymls.js /tmp/merge-latest-ymls.js
+          npm install js-yaml --prefix /tmp
+          node /tmp/merge-latest-ymls.js ./latest-mac-x64/latest-mac.yml ./latest-mac-arm64/latest-mac.yml ./latest-mac.yml
+          cat ./latest-mac.yml
+      
+      - name: Yet Another Upload Release Asset Action  
+        uses: shogo82148/actions-upload-release-asset@v1.7.2
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        with:
+          upload_url: ${{ needs.create-draft-release.outputs.upload_url }}
+          asset_path: ./latest-mac.yml
+          asset_name: latest-mac.yml
+          asset_content_type: text/yaml
+          overwrite: true
+
   update_release_draft:
-    needs: [build-macos, build-windows-x64, build-linux-x64]
+    needs: [build-macos-x64, build-macos-arm64, build-windows-x64, build-linux-x64, combine-latest-mac-yml]
     permissions:
       # write permission is required to create a github release
       contents: write

.github/workflows/template-build-macos-arm64.yml

@@ -0,0 +1,160 @@
+name: build-macos
+on:
+  workflow_call:
+    inputs:
+      ref:
+        required: true
+        type: string
+        default: 'refs/heads/main'
+      public_provider:
+        required: true
+        type: string
+        default: none
+        description: 'none: build only, github: build and publish to github, cloudflare: build and publish to cloudflare'
+      new_version:
+        required: true
+        type: string
+        default: ''
+      cloudflare_r2_path:
+        required: false
+        type: string
+        default: '/latest/'
+    secrets:
+      CLOUDFLARE_R2_BUCKET_NAME:
+        required: false
+      CLOUDFLARE_R2_ACCESS_KEY_ID:
+        required: false
+      CLOUDFLARE_R2_SECRET_ACCESS_KEY:
+        required: false
+      CLOUDFLARE_ACCOUNT_ID:
+        required: false
+      CODE_SIGN_P12_BASE64:
+        required: false
+      CODE_SIGN_P12_PASSWORD:
+        required: false
+      APPLE_ID:
+        required: false
+      APPLE_APP_SPECIFIC_PASSWORD:
+        required: false
+      DEVELOPER_ID:
+        required: false
+
+jobs:
+  build-macos:
+    runs-on: macos-silicon
+    environment: production
+    permissions:
+      contents: write
+    steps:
+      - name: Getting the repo
+        uses: actions/checkout@v3
+        with:
+          ref: ${{ inputs.ref }}
+
+      - name: Installing node
+        uses: actions/setup-node@v1
+        with:
+          node-version: 20
+      - name: Unblock keychain
+        run: |
+          security unlock-keychain -p ${{ secrets.KEYCHAIN_PASSWORD }} ~/Library/Keychains/login.keychain-db
+      # - uses: actions/setup-python@v5
+      #   with:
+      #     python-version: '3.11'
+
+      # - name: Install jq
+      #   uses: dcarbone/install-jq-action@v2.0.1
+
+      - name: Update app version based on latest release tag with build number
+        if: inputs.public_provider != 'github'
+        run: |
+          echo "Version: ${{ inputs.new_version }}"
+          # Update the version in electron/package.json
+          jq --arg version "${{ inputs.new_version }}" '.version = $version' electron/package.json > /tmp/package.json
+          mv /tmp/package.json electron/package.json
+
+          jq --arg version "${{ inputs.new_version }}" '.version = $version' web/package.json > /tmp/package.json
+          mv /tmp/package.json web/package.json
+
+          jq '.build.publish = [{"provider": "generic", "url": "${{ secrets.CLOUDFLARE_R2_PUBLIC_URL }}", "channel": "latest"}, {"provider": "s3", "bucket": "${{ secrets.CLOUDFLARE_R2_BUCKET_NAME }}", "region": "auto", "endpoint": "https://${{ secrets.CLOUDFLARE_ACCOUNT_ID }}.r2.cloudflarestorage.com", "path": "${{ inputs.cloudflare_r2_path }}", "channel": "latest"}]' electron/package.json > /tmp/package.json
+          mv /tmp/package.json electron/package.json
+          cat electron/package.json
+
+      - name: Update app version base on tag
+        if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/') && inputs.public_provider == 'github'
+        run: |
+          if [[ ! "${VERSION_TAG}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
+              echo "Error: Tag is not valid!"
+              exit 1
+          fi
+          jq --arg version "${VERSION_TAG#v}" '.version = $version' electron/package.json > /tmp/package.json
+          mv /tmp/package.json electron/package.json
+          jq --arg version "${VERSION_TAG#v}" '.version = $version' web/package.json > /tmp/package.json
+          mv /tmp/package.json web/package.json
+        env:
+          VERSION_TAG: ${{ inputs.new_version }}
+
+      # - name: Get Cer for code signing
+      #   run: base64 -d <<< "$CODE_SIGN_P12_BASE64" > /tmp/codesign.p12
+      #   shell: bash
+      #   env:
+      #     CODE_SIGN_P12_BASE64: ${{ secrets.CODE_SIGN_P12_BASE64 }}
+
+      # - uses: apple-actions/import-codesign-certs@v2
+      #   continue-on-error: true
+      #   with:
+      #     p12-file-base64: ${{ secrets.CODE_SIGN_P12_BASE64 }}
+      #     p12-password: ${{ secrets.CODE_SIGN_P12_PASSWORD }}
+
+      - name: Build and publish app to cloudflare r2 or github artifactory
+        if: inputs.public_provider != 'github'
+        run: |
+          # check public_provider is true or not
+          echo "public_provider is ${{ inputs.public_provider }}"
+          if [ "${{ inputs.public_provider }}" == "none" ]; then
+            make build    
+          else
+            make build-and-publish
+          fi
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          # CSC_LINK: "/tmp/codesign.p12"
+          # CSC_KEY_PASSWORD: ${{ secrets.CODE_SIGN_P12_PASSWORD }}
+          # CSC_IDENTITY_AUTO_DISCOVERY: "true"
+          APPLE_ID: ${{ secrets.APPLE_ID }}
+          APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
+          APP_PATH: "."
+          DEVELOPER_ID: ${{ secrets.DEVELOPER_ID }}
+          AWS_ACCESS_KEY_ID: ${{ secrets.CLOUDFLARE_R2_ACCESS_KEY_ID }}
+          AWS_SECRET_ACCESS_KEY: ${{ secrets.CLOUDFLARE_R2_SECRET_ACCESS_KEY }}
+          AWS_DEFAULT_REGION: auto
+          AWS_EC2_METADATA_DISABLED: "true"
+
+      - name: Build and publish app to github
+        if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags/') && inputs.public_provider == 'github'
+        run: |
+          make build-and-publish
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          # CSC_LINK: "/tmp/codesign.p12"
+          # CSC_KEY_PASSWORD: ${{ secrets.CODE_SIGN_P12_PASSWORD }}
+          # CSC_IDENTITY_AUTO_DISCOVERY: "true"
+          APPLE_ID: ${{ secrets.APPLE_ID }}
+          APPLE_APP_SPECIFIC_PASSWORD: ${{ secrets.APPLE_APP_SPECIFIC_PASSWORD }}
+          APP_PATH: "."
+          DEVELOPER_ID: ${{ secrets.DEVELOPER_ID }}
+          ANALYTICS_ID: ${{ secrets.JAN_APP_UMAMI_PROJECT_API_KEY }}
+          ANALYTICS_HOST: ${{ secrets.JAN_APP_UMAMI_URL }}
+
+      - name: Upload Artifact
+        if: inputs.public_provider != 'github'
+        uses: actions/upload-artifact@v2
+        with:
+          name: jan-mac-arm64-${{ inputs.new_version }}
+          path: ./electron/dist/jan-mac-arm64-${{ inputs.new_version }}.dmg
+
+      - name: Upload Artifact
+        uses: actions/upload-artifact@v2
+        with:
+          name: latest-mac-arm64
+          path: ./electron/dist/latest-mac.yml

.github/workflows/template-build-macos-x64.yml

@@ -148,9 +148,8 @@ jobs:
           path: ./electron/dist/jan-mac-x64-${{ inputs.new_version }}.dmg
 
       - name: Upload Artifact
-        if: inputs.public_provider != 'github'
         uses: actions/upload-artifact@v2
         with:
-          name: jan-mac-arm64-${{ inputs.new_version }}
+          name: latest-mac-x64
-          path: ./electron/dist/jan-mac-arm64-${{ inputs.new_version }}.dmg
+          path: ./electron/dist/latest-mac.yml
 

Dockerfile

@@ -1,39 +1,58 @@
-FROM node:20-bullseye AS base
+FROM node:20-bookworm AS base
 
 # 1. Install dependencies only when needed
-FROM base AS deps
+FROM base AS builder
+
+# Install g++ 11
+RUN apt update && apt install -y gcc-11 g++-11 cpp-11 jq xsel && rm -rf /var/lib/apt/lists/*
+
 WORKDIR /app
 
 # Install dependencies based on the preferred package manager
-COPY package.json yarn.lock* package-lock.json* pnpm-lock.yaml* ./
+COPY . ./
-RUN yarn install
+
+RUN export NITRO_VERSION=$(cat extensions/inference-nitro-extension/bin/version.txt) && \
+    jq --arg nitroVersion $NITRO_VERSION '(.scripts."downloadnitro:linux" | gsub("\\${NITRO_VERSION}"; $nitroVersion)) | gsub("\r"; "")' extensions/inference-nitro-extension/package.json > /tmp/newcommand.txt && export NEW_COMMAND=$(sed 's/^"//;s/"$//' /tmp/newcommand.txt) && jq --arg newCommand "$NEW_COMMAND" '.scripts."downloadnitro:linux" = $newCommand' extensions/inference-nitro-extension/package.json > /tmp/package.json && mv /tmp/package.json extensions/inference-nitro-extension/package.json
+RUN make install-and-build
+RUN yarn workspace jan-web install
+
+RUN export NODE_ENV=production && yarn workspace jan-web build
 
 # # 2. Rebuild the source code only when needed
-FROM base AS builder
-WORKDIR /app
-COPY --from=deps /app/node_modules ./node_modules
-COPY . .
-# This will do the trick, use the corresponding env file for each environment.
-RUN yarn workspace server install
-RUN yarn server:prod
-
-# 3. Production image, copy all the files and run next
 FROM base AS runner
+
+# Install g++ 11
+RUN apt update && apt install -y gcc-11 g++-11 cpp-11 jq xsel && rm -rf /var/lib/apt/lists/*
+
 WORKDIR /app
 
-ENV NODE_ENV=production
+# Copy the package.json and yarn.lock of root yarn space to leverage Docker cache
+COPY --from=builder /app/package.json ./package.json
+COPY --from=builder /app/node_modules ./node_modules/
+COPY --from=builder /app/yarn.lock ./yarn.lock
 
-# RUN addgroup -g 1001 -S nodejs;
+# Copy the package.json, yarn.lock, and build output of server yarn space to leverage Docker cache
-COPY --from=builder /app/server/build ./
+COPY --from=builder /app/server ./server/
+COPY --from=builder /app/docs/openapi ./docs/openapi/
 
-# Automatically leverage output traces to reduce image size
+# Copy pre-install dependencies
-# https://nextjs.org/docs/advanced-features/output-file-tracing
+COPY --from=builder /app/pre-install ./pre-install/
-COPY --from=builder /app/server/node_modules ./node_modules
-COPY --from=builder /app/server/package.json ./package.json
 
-EXPOSE 4000 3928
+# Copy the package.json, yarn.lock, and output of web yarn space to leverage Docker cache
+COPY --from=builder /app/web/out ./web/out/
+COPY --from=builder /app/web/.next ./web/.next/
+COPY --from=builder /app/web/package.json ./web/package.json
+COPY --from=builder /app/web/yarn.lock ./web/yarn.lock
+COPY --from=builder /app/models ./models/
 
-ENV PORT 4000
+RUN npm install -g serve@latest
-ENV APPDATA /app/data
 
-CMD ["node", "main.js"]
+EXPOSE 1337 3000 3928
+
+ENV JAN_API_HOST 0.0.0.0
+ENV JAN_API_PORT 1337
+
+CMD ["sh", "-c", "cd server && node build/main.js & cd web && npx serve out"]
+
+# docker build -t jan .
+# docker run -p 1337:1337 -p 3000:3000 -p 3928:3928 jan

Dockerfile.gpu

@@ -0,0 +1,85 @@
+# Please change the base image to the appropriate CUDA version base on NVIDIA Driver Compatibility
+# Run nvidia-smi to check the CUDA version and the corresponding driver version
+# Then update the base image to the appropriate CUDA version refer https://catalog.ngc.nvidia.com/orgs/nvidia/containers/cuda/tags
+
+FROM nvidia/cuda:12.2.0-runtime-ubuntu22.04 AS base 
+
+# 1. Install dependencies only when needed
+FROM base AS builder
+
+# Install g++ 11
+RUN apt update && apt install -y gcc-11 g++-11 cpp-11 jq xsel curl gnupg make python3-dev && curl -sL https://deb.nodesource.com/setup_20.x | bash - && apt install nodejs -y && rm -rf /var/lib/apt/lists/*
+
+# Update alternatives for GCC and related tools
+RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 110 \
+                         --slave /usr/bin/g++ g++ /usr/bin/g++-11 \
+                         --slave /usr/bin/gcov gcov /usr/bin/gcov-11 \
+                         --slave /usr/bin/gcc-ar gcc-ar /usr/bin/gcc-ar-11 \
+                         --slave /usr/bin/gcc-ranlib gcc-ranlib /usr/bin/gcc-ranlib-11 && \
+    update-alternatives --install /usr/bin/cpp cpp /usr/bin/cpp-11 110
+
+RUN npm install -g yarn
+
+WORKDIR /app
+
+# Install dependencies based on the preferred package manager
+COPY . ./
+
+RUN export NITRO_VERSION=$(cat extensions/inference-nitro-extension/bin/version.txt) && \
+    jq --arg nitroVersion $NITRO_VERSION '(.scripts."downloadnitro:linux" | gsub("\\${NITRO_VERSION}"; $nitroVersion)) | gsub("\r"; "")' extensions/inference-nitro-extension/package.json > /tmp/newcommand.txt && export NEW_COMMAND=$(sed 's/^"//;s/"$//' /tmp/newcommand.txt) && jq --arg newCommand "$NEW_COMMAND" '.scripts."downloadnitro:linux" = $newCommand' extensions/inference-nitro-extension/package.json > /tmp/package.json && mv /tmp/package.json extensions/inference-nitro-extension/package.json
+RUN make install-and-build
+RUN yarn workspace jan-web install
+
+RUN export NODE_ENV=production && yarn workspace jan-web build
+
+# # 2. Rebuild the source code only when needed
+FROM base AS runner
+
+# Install g++ 11
+RUN apt update && apt install -y gcc-11 g++-11 cpp-11 jq xsel curl gnupg make python3-dev && curl -sL https://deb.nodesource.com/setup_20.x | bash - && apt-get install nodejs -y && rm -rf /var/lib/apt/lists/*
+
+# Update alternatives for GCC and related tools
+RUN update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-11 110 \
+                         --slave /usr/bin/g++ g++ /usr/bin/g++-11 \
+                         --slave /usr/bin/gcov gcov /usr/bin/gcov-11 \
+                         --slave /usr/bin/gcc-ar gcc-ar /usr/bin/gcc-ar-11 \
+                         --slave /usr/bin/gcc-ranlib gcc-ranlib /usr/bin/gcc-ranlib-11 && \
+    update-alternatives --install /usr/bin/cpp cpp /usr/bin/cpp-11 110
+
+RUN npm install -g yarn
+
+WORKDIR /app
+
+# Copy the package.json and yarn.lock of root yarn space to leverage Docker cache
+COPY --from=builder /app/package.json ./package.json
+COPY --from=builder /app/node_modules ./node_modules/
+COPY --from=builder /app/yarn.lock ./yarn.lock
+
+# Copy the package.json, yarn.lock, and build output of server yarn space to leverage Docker cache
+COPY --from=builder /app/server ./server/
+COPY --from=builder /app/docs/openapi ./docs/openapi/
+
+# Copy pre-install dependencies
+COPY --from=builder /app/pre-install ./pre-install/
+
+# Copy the package.json, yarn.lock, and output of web yarn space to leverage Docker cache
+COPY --from=builder /app/web/out ./web/out/
+COPY --from=builder /app/web/.next ./web/.next/
+COPY --from=builder /app/web/package.json ./web/package.json
+COPY --from=builder /app/web/yarn.lock ./web/yarn.lock
+COPY --from=builder /app/models ./models/
+
+RUN npm install -g serve@latest
+
+EXPOSE 1337 3000 3928
+
+ENV LD_LIBRARY_PATH=/usr/local/cuda/targets/x86_64-linux/lib:/usr/local/cuda-12.0/compat${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
+
+ENV JAN_API_HOST 0.0.0.0
+ENV JAN_API_PORT 1337
+
+CMD ["sh", "-c", "cd server && node build/main.js & cd web && npx serve out"]
+
+# pre-requisites: nvidia-docker
+# docker build -t jan-gpu . -f Dockerfile.gpu
+# docker run -p 1337:1337 -p 3000:3000 -p 3928:3928 --gpus all jan-gpu

Makefile

@@ -24,9 +24,9 @@ endif
 
 check-file-counts: install-and-build
 ifeq ($(OS),Windows_NT)
-	powershell -Command "if ((Get-ChildItem -Path electron/pre-install -Filter *.tgz | Measure-Object | Select-Object -ExpandProperty Count) -ne (Get-ChildItem -Path extensions -Directory | Measure-Object | Select-Object -ExpandProperty Count)) { Write-Host 'Number of .tgz files in electron/pre-install does not match the number of subdirectories in extension'; exit 1 } else { Write-Host 'Extension build successful' }"
+	powershell -Command "if ((Get-ChildItem -Path pre-install -Filter *.tgz | Measure-Object | Select-Object -ExpandProperty Count) -ne (Get-ChildItem -Path extensions -Directory | Measure-Object | Select-Object -ExpandProperty Count)) { Write-Host 'Number of .tgz files in pre-install does not match the number of subdirectories in extension'; exit 1 } else { Write-Host 'Extension build successful' }"
 else
-	@tgz_count=$$(find electron/pre-install -type f -name "*.tgz" | wc -l); dir_count=$$(find extensions -mindepth 1 -maxdepth 1 -type d | wc -l); if [ $$tgz_count -ne $$dir_count ]; then echo "Number of .tgz files in electron/pre-install ($$tgz_count) does not match the number of subdirectories in extension ($$dir_count)"; exit 1; else echo "Extension build successful"; fi
+	@tgz_count=$$(find pre-install -type f -name "*.tgz" | wc -l); dir_count=$$(find extensions -mindepth 1 -maxdepth 1 -type d | wc -l); if [ $$tgz_count -ne $$dir_count ]; then echo "Number of .tgz files in pre-install ($$tgz_count) does not match the number of subdirectories in extension ($$dir_count)"; exit 1; else echo "Extension build successful"; fi
 endif
 
 dev: check-file-counts

README.md

@@ -43,31 +43,31 @@ Jan is an open-source ChatGPT alternative that runs 100% offline on your compute
   <tr style="text-align:center">
     <td style="text-align:center"><b>Stable (Recommended)</b></td>
     <td style="text-align:center">
-      <a href='https://github.com/janhq/jan/releases/download/v0.4.5/jan-win-x64-0.4.5.exe'>
+      <a href='https://github.com/janhq/jan/releases/download/v0.4.6/jan-win-x64-0.4.6.exe'>
         <img src='./docs/static/img/windows.png' style="height:14px; width: 14px" />
         <b>jan.exe</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://github.com/janhq/jan/releases/download/v0.4.5/jan-mac-x64-0.4.5.dmg'>
+      <a href='https://github.com/janhq/jan/releases/download/v0.4.6/jan-mac-x64-0.4.6.dmg'>
         <img src='./docs/static/img/mac.png' style="height:15px; width: 15px" />
         <b>Intel</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://github.com/janhq/jan/releases/download/v0.4.5/jan-mac-arm64-0.4.5.dmg'>
+      <a href='https://github.com/janhq/jan/releases/download/v0.4.6/jan-mac-arm64-0.4.6.dmg'>
         <img src='./docs/static/img/mac.png' style="height:15px; width: 15px" />
         <b>M1/M2</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://github.com/janhq/jan/releases/download/v0.4.5/jan-linux-amd64-0.4.5.deb'>
+      <a href='https://github.com/janhq/jan/releases/download/v0.4.6/jan-linux-amd64-0.4.6.deb'>
         <img src='./docs/static/img/linux.png' style="height:14px; width: 14px" />
         <b>jan.deb</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://github.com/janhq/jan/releases/download/v0.4.5/jan-linux-x86_64-0.4.5.AppImage'>
+      <a href='https://github.com/janhq/jan/releases/download/v0.4.6/jan-linux-x86_64-0.4.6.AppImage'>
         <img src='./docs/static/img/linux.png' style="height:14px; width: 14px" />
         <b>jan.AppImage</b>
       </a>
@@ -76,31 +76,31 @@ Jan is an open-source ChatGPT alternative that runs 100% offline on your compute
   <tr style="text-align:center">
     <td style="text-align:center"><b>Experimental (Nightly Build)</b></td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-win-x64-0.4.5-260.exe'>
+      <a href='https://delta.jan.ai/latest/jan-win-x64-0.4.6-264.exe'>
         <img src='./docs/static/img/windows.png' style="height:14px; width: 14px" />
         <b>jan.exe</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-mac-x64-0.4.5-260.dmg'>
+      <a href='https://delta.jan.ai/latest/jan-mac-x64-0.4.6-264.dmg'>
         <img src='./docs/static/img/mac.png' style="height:15px; width: 15px" />
         <b>Intel</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-mac-arm64-0.4.5-260.dmg'>
+      <a href='https://delta.jan.ai/latest/jan-mac-arm64-0.4.6-264.dmg'>
         <img src='./docs/static/img/mac.png' style="height:15px; width: 15px" />
         <b>M1/M2</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-linux-amd64-0.4.5-260.deb'>
+      <a href='https://delta.jan.ai/latest/jan-linux-amd64-0.4.6-264.deb'>
         <img src='./docs/static/img/linux.png' style="height:14px; width: 14px" />
         <b>jan.deb</b>
       </a>
     </td>
     <td style="text-align:center">
-      <a href='https://delta.jan.ai/latest/jan-linux-x86_64-0.4.5-260.AppImage'>
+      <a href='https://delta.jan.ai/latest/jan-linux-x86_64-0.4.6-264.AppImage'>
         <img src='./docs/static/img/linux.png' style="height:14px; width: 14px" />
         <b>jan.AppImage</b>
       </a>
@@ -218,6 +218,76 @@ make build
 
 This will build the app MacOS m1/m2 for production (with code signing already done) and put the result in `dist` folder.
 
+### Docker mode
+
+- Supported OS: Linux, WSL2 Docker
+- Pre-requisites:
+  - `docker` and `docker compose`, follow instruction [here](https://docs.docker.com/engine/install/ubuntu/)
+
+    ```bash
+    curl -fsSL https://get.docker.com -o get-docker.sh
+    sudo sh ./get-docker.sh --dry-run
+    ```
+
+  - `nvidia-driver` and `nvidia-docker2`, follow instruction [here](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html) (If you want to run with GPU mode)
+
+- Run Jan in Docker mode
+
+  - **Option 1**: Run Jan in CPU mode
+
+    ```bash
+    docker compose --profile cpu up -d
+    ```
+
+  - **Option 2**: Run Jan in GPU mode
+
+    - **Step 1**: Check cuda compatibility with your nvidia driver by running `nvidia-smi` and check the cuda version in the output
+
+      ```bash
+      nvidia-smi
+
+      # Output
+      +---------------------------------------------------------------------------------------+
+      | NVIDIA-SMI 531.18                 Driver Version: 531.18       CUDA Version: 12.1     |
+      |-----------------------------------------+----------------------+----------------------+
+      | GPU  Name                      TCC/WDDM | Bus-Id        Disp.A | Volatile Uncorr. ECC |
+      | Fan  Temp  Perf            Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
+      |                                         |                      |               MIG M. |
+      |=========================================+======================+======================|
+      |   0  NVIDIA GeForce RTX 4070 Ti    WDDM | 00000000:01:00.0  On |                  N/A |
+      |  0%   44C    P8               16W / 285W|   1481MiB / 12282MiB |      2%      Default |
+      |                                         |                      |                  N/A |
+      +-----------------------------------------+----------------------+----------------------+
+      |   1  NVIDIA GeForce GTX 1660 Ti    WDDM | 00000000:02:00.0 Off |                  N/A |
+      |  0%   49C    P8               14W / 120W|      0MiB /  6144MiB |      0%      Default |
+      |                                         |                      |                  N/A |
+      +-----------------------------------------+----------------------+----------------------+
+      |   2  NVIDIA GeForce GTX 1660 Ti    WDDM | 00000000:05:00.0 Off |                  N/A |
+      | 29%   38C    P8               11W / 120W|      0MiB /  6144MiB |      0%      Default |
+      |                                         |                      |                  N/A |
+      +-----------------------------------------+----------------------+----------------------+
+
+      +---------------------------------------------------------------------------------------+
+      | Processes:                                                                            |
+      |  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
+      |        ID   ID                                                             Usage      |
+      |=======================================================================================|
+      ```
+
+    - **Step 2**: Go to https://catalog.ngc.nvidia.com/orgs/nvidia/containers/cuda/tags and find the smallest minor version of image tag that matches the cuda version from the output of `nvidia-smi` (e.g. 12.1 -> 12.1.0)
+
+    - **Step 3**: Update the `Dockerfile.gpu` line number 5 with the latest minor version of the image tag from step 2 (e.g. change `FROM nvidia/cuda:12.2.0-runtime-ubuntu22.04 AS base` to `FROM nvidia/cuda:12.1.0-runtime-ubuntu22.04 AS base`)
+
+    - **Step 4**: Run command to start Jan in GPU mode
+
+      ```bash
+      # GPU mode
+      docker compose --profile gpu up -d
+      ```
+
+  This will start the web server and you can access Jan at `http://localhost:3000`.
+  > Note: Currently, Docker mode is only work for development and localhost, production is not supported yet. RAG feature is not supported in Docker mode yet.
+
 ## Acknowledgements
 
 Jan builds on top of other open-source projects:

core/package.json

@@ -57,6 +57,7 @@
     "rollup-plugin-typescript2": "^0.36.0",
     "ts-jest": "^26.1.1",
     "tslib": "^2.6.2",
-    "typescript": "^5.2.2"
+    "typescript": "^5.2.2",
+    "rimraf": "^3.0.2"
   }
 }

core/src/api/index.ts

@@ -30,6 +30,7 @@ export enum DownloadRoute {
   downloadFile = 'downloadFile',
   pauseDownload = 'pauseDownload',
   resumeDownload = 'resumeDownload',
+  getDownloadProgress = 'getDownloadProgress',
 }
 
 export enum DownloadEvent {

core/src/node/api/routes/common.ts

@@ -12,6 +12,8 @@ import {
 import { JanApiRouteConfiguration } from '../common/configuration'
 import { startModel, stopModel } from '../common/startStopModel'
 import { ModelSettingParams } from '../../../types'
+import { getJanDataFolderPath } from '../../utils'
+import { normalizeFilePath } from '../../path'
 
 export const commonRouter = async (app: HttpServer) => {
   // Common Routes
@@ -52,7 +54,14 @@ export const commonRouter = async (app: HttpServer) => {
   // App Routes
   app.post(`/app/${AppRoute.joinPath}`, async (request: any, reply: any) => {
     const args = JSON.parse(request.body) as any[]
-    reply.send(JSON.stringify(join(...args[0])))
+
+    const paths = args[0].map((arg: string) =>
+      typeof arg === 'string' && (arg.startsWith(`file:/`) || arg.startsWith(`file:\\`))
+        ? join(getJanDataFolderPath(), normalizeFilePath(arg))
+        : arg
+    )
+
+    reply.send(JSON.stringify(join(...paths)))
   })
 
   app.post(`/app/${AppRoute.baseName}`, async (request: any, reply: any) => {

core/src/node/api/routes/download.ts

@@ -4,55 +4,109 @@ import { DownloadManager } from '../../download'
 import { HttpServer } from '../HttpServer'
 import { createWriteStream } from 'fs'
 import { getJanDataFolderPath } from '../../utils'
-import { normalizeFilePath } from "../../path";
+import { normalizeFilePath } from '../../path'
+import { DownloadState } from '../../../types'
 
 export const downloadRouter = async (app: HttpServer) => {
+  app.get(`/${DownloadRoute.getDownloadProgress}/:modelId`, async (req, res) => {
+    const modelId = req.params.modelId
+
+    console.debug(`Getting download progress for model ${modelId}`)
+    console.debug(
+      `All Download progress: ${JSON.stringify(DownloadManager.instance.downloadProgressMap)}`
+    )
+
+    // check if null DownloadManager.instance.downloadProgressMap
+    if (!DownloadManager.instance.downloadProgressMap[modelId]) {
+      return res.status(404).send({
+        message: 'Download progress not found',
+      })
+    } else {
+      return res.status(200).send(DownloadManager.instance.downloadProgressMap[modelId])
+    }
+  })
+
   app.post(`/${DownloadRoute.downloadFile}`, async (req, res) => {
-    const strictSSL = !(req.query.ignoreSSL === "true");
+    const strictSSL = !(req.query.ignoreSSL === 'true')
-    const proxy = req.query.proxy?.startsWith("http") ? req.query.proxy : undefined;
+    const proxy = req.query.proxy?.startsWith('http') ? req.query.proxy : undefined
-    const body = JSON.parse(req.body as any);
+    const body = JSON.parse(req.body as any)
     const normalizedArgs = body.map((arg: any) => {
-      if (typeof arg === "string") {
+      if (typeof arg === 'string' && arg.startsWith('file:')) {
-        return join(getJanDataFolderPath(), normalizeFilePath(arg));
+        return join(getJanDataFolderPath(), normalizeFilePath(arg))
       }
-      return arg;
+      return arg
-    });
+    })
 
-    const localPath = normalizedArgs[1];
+    const localPath = normalizedArgs[1]
-    const fileName = localPath.split("/").pop() ?? "";
+    const array = localPath.split('/')
+    const fileName = array.pop() ?? ''
+    const modelId = array.pop() ?? ''
+    console.debug('downloadFile', normalizedArgs, fileName, modelId)
 
-    const request = require("request");
+    const request = require('request')
-    const progress = require("request-progress");
+    const progress = require('request-progress')
 
-    const rq = request({ url: normalizedArgs[0], strictSSL, proxy });
+    const rq = request({ url: normalizedArgs[0], strictSSL, proxy })
     progress(rq, {})
-      .on("progress", function (state: any) {
+      .on('progress', function (state: any) {
-        console.log("download onProgress", state);
+        const downloadProps: DownloadState = {
+          ...state,
+          modelId,
+          fileName,
+          downloadState: 'downloading',
+        }
+        console.debug(`Download ${modelId} onProgress`, downloadProps)
+        DownloadManager.instance.downloadProgressMap[modelId] = downloadProps
       })
-      .on("error", function (err: Error) {
+      .on('error', function (err: Error) {
-        console.log("download onError", err);
+        console.debug(`Download ${modelId} onError`, err.message)
-      })
-      .on("end", function () {
-        console.log("download onEnd");
-      })
-      .pipe(createWriteStream(normalizedArgs[1]));
 
-    DownloadManager.instance.setRequest(fileName, rq);
+        const currentDownloadState = DownloadManager.instance.downloadProgressMap[modelId]
-  });
+        if (currentDownloadState) {
+          DownloadManager.instance.downloadProgressMap[modelId] = {
+            ...currentDownloadState,
+            downloadState: 'error',
+          }
+        }
+      })
+      .on('end', function () {
+        console.debug(`Download ${modelId} onEnd`)
+
+        const currentDownloadState = DownloadManager.instance.downloadProgressMap[modelId]
+        if (currentDownloadState) {
+          if (currentDownloadState.downloadState === 'downloading') {
+            // if the previous state is downloading, then set the state to end (success)
+            DownloadManager.instance.downloadProgressMap[modelId] = {
+              ...currentDownloadState,
+              downloadState: 'end',
+            }
+          }
+        }
+      })
+      .pipe(createWriteStream(normalizedArgs[1]))
+
+    DownloadManager.instance.setRequest(localPath, rq)
+    res.status(200).send({ message: 'Download started' })
+  })
 
   app.post(`/${DownloadRoute.abortDownload}`, async (req, res) => {
-    const body = JSON.parse(req.body as any);
+    const body = JSON.parse(req.body as any)
     const normalizedArgs = body.map((arg: any) => {
-      if (typeof arg === "string") {
+      if (typeof arg === 'string' && arg.startsWith('file:')) {
-        return join(getJanDataFolderPath(), normalizeFilePath(arg));
+        return join(getJanDataFolderPath(), normalizeFilePath(arg))
       }
-      return arg;
+      return arg
-    });
+    })
 
-    const localPath = normalizedArgs[0];
+    const localPath = normalizedArgs[0]
-    const fileName = localPath.split("/").pop() ?? "";
+    const fileName = localPath.split('/').pop() ?? ''
-    const rq = DownloadManager.instance.networkRequests[fileName];
+    const rq = DownloadManager.instance.networkRequests[fileName]
-    DownloadManager.instance.networkRequests[fileName] = undefined;
+    DownloadManager.instance.networkRequests[fileName] = undefined
-    rq?.abort();
+    rq?.abort()
-  });
+    if (rq) {
-};
+      res.status(200).send({ message: 'Download aborted' })
+    } else {
+      res.status(404).send({ message: 'Download not found' })
+    }
+  })
+}

core/src/node/api/routes/fileManager.ts

@@ -1,14 +1,29 @@
 import { FileManagerRoute } from '../../../api'
 import { HttpServer } from '../../index'
+import { join } from 'path'
 
-export const fsRouter = async (app: HttpServer) => {
+export const fileManagerRouter = async (app: HttpServer) => {
-  app.post(`/app/${FileManagerRoute.syncFile}`, async (request: any, reply: any) => {})
+  app.post(`/fs/${FileManagerRoute.syncFile}`, async (request: any, reply: any) => {
+    const reflect = require('@alumna/reflect')
+    const args = JSON.parse(request.body)
+    return reflect({
+      src: args[0],
+      dest: args[1],
+      recursive: true,
+      delete: false,
+      overwrite: true,
+      errorOnExist: false,
+    })
+  })
 
-  app.post(`/app/${FileManagerRoute.getJanDataFolderPath}`, async (request: any, reply: any) => {})
+  app.post(`/fs/${FileManagerRoute.getJanDataFolderPath}`, async (request: any, reply: any) =>
+    global.core.appPath()
+  )
 
-  app.post(`/app/${FileManagerRoute.getResourcePath}`, async (request: any, reply: any) => {})
+  app.post(`/fs/${FileManagerRoute.getResourcePath}`, async (request: any, reply: any) =>
+    join(global.core.appPath(), '../../..')
+  )
 
   app.post(`/app/${FileManagerRoute.getUserHomePath}`, async (request: any, reply: any) => {})
-
+  app.post(`/fs/${FileManagerRoute.fileStat}`, async (request: any, reply: any) => {})
-  app.post(`/app/${FileManagerRoute.fileStat}`, async (request: any, reply: any) => {})
 }

core/src/node/api/routes/fs.ts

@@ -1,8 +1,9 @@
-import { FileSystemRoute } from '../../../api'
+import { FileManagerRoute, FileSystemRoute } from '../../../api'
 import { join } from 'path'
 import { HttpServer } from '../HttpServer'
 import { getJanDataFolderPath } from '../../utils'
 import { normalizeFilePath } from '../../path'
+import { writeFileSync } from 'fs'
 
 export const fsRouter = async (app: HttpServer) => {
   const moduleName = 'fs'
@@ -26,4 +27,14 @@ export const fsRouter = async (app: HttpServer) => {
       }
     })
   })
+  app.post(`/${FileManagerRoute.writeBlob}`, async (request: any, reply: any) => {
+    try {
+      const args = JSON.parse(request.body) as any[]
+      console.log('writeBlob:', args[0])
+      const dataBuffer = Buffer.from(args[1], 'base64')
+      writeFileSync(args[0], dataBuffer)
+    } catch (err) {
+      console.error(`writeFile ${request.body} result: ${err}`)
+    }
+  })
 }

core/src/node/api/routes/v1.ts

@@ -4,6 +4,7 @@ import { threadRouter } from './thread'
 import { fsRouter } from './fs'
 import { extensionRouter } from './extension'
 import { downloadRouter } from './download'
+import { fileManagerRouter } from './fileManager'
 
 export const v1Router = async (app: HttpServer) => {
   // MARK: External Routes
@@ -16,6 +17,8 @@ export const v1Router = async (app: HttpServer) => {
   app.register(fsRouter, {
     prefix: '/fs',
   })
+  app.register(fileManagerRouter)
+
   app.register(extensionRouter, {
     prefix: '/extension',
   })

core/src/node/download.ts

@@ -1,15 +1,18 @@
+import { DownloadState } from '../types'
 
 /**
  * Manages file downloads and network requests.
  */
 export class DownloadManager {
-  public networkRequests: Record<string, any> = {};
+  public networkRequests: Record<string, any> = {}
 
-  public static instance: DownloadManager = new DownloadManager();
+  public static instance: DownloadManager = new DownloadManager()
+
+  public downloadProgressMap: Record<string, DownloadState> = {}
 
   constructor() {
     if (DownloadManager.instance) {
-      return DownloadManager.instance;
+      return DownloadManager.instance
     }
   }
   /**
@@ -18,6 +21,6 @@ export class DownloadManager {
    * @param {Request | undefined} request - The network request to set, or undefined to clear the request.
    */
   setRequest(fileName: string, request: any | undefined) {
-    this.networkRequests[fileName] = request;
+    this.networkRequests[fileName] = request
   }
 }

core/src/node/extension/index.ts

@@ -41,8 +41,8 @@ async function registerExtensionProtocol() {
     console.error('Electron is not available')
   }
   const extensionPath = ExtensionManager.instance.getExtensionsPath()
-  if (electron) {
+  if (electron && electron.protocol) {
-    return electron.protocol.registerFileProtocol('extension', (request: any, callback: any) => {
+    return electron.protocol?.registerFileProtocol('extension', (request: any, callback: any) => {
       const entry = request.url.substr('extension://'.length - 1)
 
       const url = normalize(extensionPath + entry)
@@ -69,7 +69,7 @@ export function useExtensions(extensionsPath: string) {
 
   // Read extension list from extensions folder
   const extensions = JSON.parse(
-    readFileSync(ExtensionManager.instance.getExtensionsFile(), 'utf-8'),
+    readFileSync(ExtensionManager.instance.getExtensionsFile(), 'utf-8')
   )
   try {
     // Create and store a Extension instance for each extension in list
@@ -82,7 +82,7 @@ export function useExtensions(extensionsPath: string) {
     throw new Error(
       'Could not successfully rebuild list of installed extensions.\n' +
         error +
-        '\nPlease check the extensions.json file in the extensions folder.',
+        '\nPlease check the extensions.json file in the extensions folder.'
     )
   }
 
@@ -122,7 +122,7 @@ function loadExtension(ext: any) {
 export function getStore() {
   if (!ExtensionManager.instance.getExtensionsFile()) {
     throw new Error(
-      'The extension path has not yet been set up. Please run useExtensions before accessing the store',
+      'The extension path has not yet been set up. Please run useExtensions before accessing the store'
     )
   }
 

core/src/types/assistant/assistantEvent.ts

@@ -0,0 +1,8 @@
+/**
+ * The `EventName` enumeration contains the names of all the available events in the Jan platform.
+ */
+export enum AssistantEvent {
+    /** The `OnAssistantsUpdate` event is emitted when the assistant list is updated. */
+    OnAssistantsUpdate = 'OnAssistantsUpdate',
+  }
+  

core/src/types/assistant/index.ts

@@ -1,2 +1,3 @@
 export * from './assistantEntity'
+export * from './assistantEvent'
 export * from './assistantInterface'

core/src/types/file/index.ts

@@ -2,3 +2,26 @@ export type FileStat = {
   isDirectory: boolean
   size: number
 }
+
+export type DownloadState = {
+  modelId: string
+  filename: string
+  time: DownloadTime
+  speed: number
+  percent: number
+
+  size: DownloadSize
+  children?: DownloadState[]
+  error?: string
+  downloadState: 'downloading' | 'error' | 'end'
+}
+
+type DownloadTime = {
+  elapsed: number
+  remaining: number
+}
+
+type DownloadSize = {
+  total: number
+  transferred: number
+}

core/src/types/model/modelEvent.ts

@@ -12,4 +12,6 @@ export enum ModelEvent {
   OnModelStop = 'OnModelStop',
   /** The `OnModelStopped` event is emitted when a model stopped ok. */
   OnModelStopped = 'OnModelStopped',
+  /** The `OnModelUpdate` event is emitted when the model list is updated. */
+  OnModelsUpdate = 'OnModelsUpdate',
 }

docker-compose.yml

@@ -0,0 +1,117 @@
+# Docker Compose file for setting up Minio, createbuckets, app_cpu, and app_gpu services
+
+version: '3.7'
+
+services:
+  # Minio service for object storage
+  minio:
+    image: minio/minio
+    volumes:
+      - minio_data:/data
+    ports:
+      - "9000:9000"
+      - "9001:9001"
+    environment:
+      # Set the root user and password for Minio
+      MINIO_ROOT_USER: minioadmin # This acts as AWS_ACCESS_KEY
+      MINIO_ROOT_PASSWORD: minioadmin # This acts as AWS_SECRET_ACCESS_KEY
+    command: server --console-address ":9001" /data
+    restart: always
+    healthcheck:
+      test: ["CMD", "curl", "-f", "http://localhost:9000/minio/health/live"]
+      interval: 30s
+      timeout: 20s
+      retries: 3
+    networks:
+      vpcbr:
+        ipv4_address: 10.5.0.2
+
+  # createbuckets service to create a bucket and set its policy
+  createbuckets:
+    image: minio/mc
+    depends_on:
+      - minio
+    entrypoint: >
+      /bin/sh -c "
+      /usr/bin/mc alias set myminio http://minio:9000 minioadmin minioadmin;
+      /usr/bin/mc mb myminio/mybucket;
+      /usr/bin/mc policy set public myminio/mybucket;
+      exit 0;
+      "
+    networks:
+      vpcbr:
+
+  # app_cpu service for running the CPU version of the application
+  app_cpu:
+    image: jan:latest
+    volumes:
+      - app_data:/app/server/build/jan
+    build:
+      context: .
+      dockerfile: Dockerfile
+    environment:
+      # Set the AWS access key, secret access key, bucket name, endpoint, and region for app_cpu
+      AWS_ACCESS_KEY_ID: minioadmin
+      AWS_SECRET_ACCESS_KEY: minioadmin
+      S3_BUCKET_NAME: mybucket
+      AWS_ENDPOINT: http://10.5.0.2:9000
+      AWS_REGION: us-east-1
+    restart: always
+    profiles:
+      - cpu
+    ports:
+      - "3000:3000"
+      - "1337:1337"
+      - "3928:3928"
+    networks:
+      vpcbr:
+        ipv4_address: 10.5.0.3
+
+  # app_gpu service for running the GPU version of the application
+  app_gpu:
+    deploy:
+      resources:
+        reservations:
+          devices:
+          - driver: nvidia
+            count: all
+            capabilities: [gpu]
+    image: jan-gpu:latest
+    volumes:
+      - app_data:/app/server/build/jan
+    build:
+      context: .
+      dockerfile: Dockerfile.gpu
+    restart: always
+    environment:
+      # Set the AWS access key, secret access key, bucket name, endpoint, and region for app_gpu
+      AWS_ACCESS_KEY_ID: minioadmin
+      AWS_SECRET_ACCESS_KEY: minioadmin
+      S3_BUCKET_NAME: mybucket
+      AWS_ENDPOINT: http://10.5.0.2:9000
+      AWS_REGION: us-east-1
+    profiles:
+      - gpu
+    ports:
+      - "3000:3000"
+      - "1337:1337"
+      - "3928:3928"
+    networks:
+      vpcbr:
+        ipv4_address: 10.5.0.4
+
+volumes:
+  minio_data:
+  app_data:
+
+networks:
+  vpcbr:
+    driver: bridge
+    ipam:
+     config:
+       - subnet: 10.5.0.0/16
+         gateway: 10.5.0.1
+
+# Usage:
+# - Run 'docker-compose --profile cpu up -d' to start the app_cpu service
+# - Run 'docker-compose --profile gpu up -d' to start the app_gpu service

docs/blog/authors.yml

@@ -4,3 +4,60 @@ dan-jan:
   url: https://github.com/dan-jan
   image_url: https://avatars.githubusercontent.com/u/101145494?v=4
   email: daniel@jan.ai
+
+namchuai:
+  name: Nam Nguyen
+  title: Developer
+  url: https://github.com/namchuai
+  image_url: https://avatars.githubusercontent.com/u/10397206?v=4
+  email: james@jan.ai
+  
+hiro-v:
+  name: Hiro Vuong
+  title: MLE
+  url: https://github.com/hiro-v
+  image_url: https://avatars.githubusercontent.com/u/22463238?v=4
+  email: hiro@jan.ai
+
+ashley-jan:
+  name: Ashley Tran
+  title: Product Designer
+  url: https://github.com/imtuyethan
+  image_url: https://avatars.githubusercontent.com/u/89722390?v=4
+  email: ashley@jan.ai
+
+hientominh:
+  name: Hien To
+  title: DevOps Engineer
+  url: https://github.com/hientominh
+  image_url: https://avatars.githubusercontent.com/u/37921427?v=4
+  email: hien@jan.ai
+
+Van-QA:
+  name: Van Pham
+  title: QA & Release Manager
+  url: https://github.com/Van-QA
+  image_url: https://avatars.githubusercontent.com/u/64197333?v=4
+  email: van@jan.ai
+
+louis-jan:
+  name: Louis Le
+  title: Software Engineer
+  url: https://github.com/louis-jan
+  image_url: https://avatars.githubusercontent.com/u/133622055?v=4
+  email: louis@jan.ai
+
+hahuyhoang411:
+  name: Rex Ha
+  title: LLM Researcher & Content Writer
+  url: https://github.com/hahuyhoang411
+  image_url: https://avatars.githubusercontent.com/u/64120343?v=4
+  email: rex@jan.ai
+
+automaticcat:
+  name: Alan Dao
+  title: AI Engineer
+  url: https://github.com/tikikun
+  image_url: https://avatars.githubusercontent.com/u/22268502?v=4
+  email: alan@jan.ai
+  

docs/docs/about/01-README.md

@@ -110,9 +110,10 @@ Adhering to Jan's privacy preserving philosophy, our analytics philosophy is to
 
 #### What is tracked
 
-1. By default, Github tracks downloads and device metadata for all public Github repos. This helps us troubleshoot & ensure cross platform support.
+1. By default, Github tracks downloads and device metadata for all public GitHub repositories. This helps us troubleshoot & ensure cross-platform support.
-1. We use Posthog to track a single `app.opened` event without additional user metadata, in order to understand retention.
+2. We use [Umami](https://umami.is/) to collect, analyze, and understand application data while maintaining visitor privacy and data ownership. We are using the Umami Cloud in Europe to ensure GDPR compliance. Please see [Umami Privacy Policy](https://umami.is/privacy) for more details.
-1. Additionally, we plan to enable a `Settings` feature for users to turn off all tracking.
+3. We use Umami to track a single `app.opened` event without additional user metadata, in order to understand retention. In addition, we track `app.event` to understand app version usage.
+4. Additionally, we plan to enable a `Settings` feature for users to turn off all tracking.
 
 #### Request for help
 

docs/docs/developer/01-overview/04-install-and-prerequisites.md

@@ -0,0 +1,79 @@
+---
+title: Installation and Prerequisites
+slug: /developer/prereq
+description: Guide to install and setup Jan for development.
+keywords:
+  [
+    Jan AI,
+    Jan,
+    ChatGPT alternative,
+    local AI,
+    private AI,
+    conversational AI,
+    no-subscription fee,
+    large language model,
+    installation,
+    prerequisites,
+    developer setup,
+  ]
+---
+
+## Requirements
+
+### Hardware Requirements
+
+Ensure your system meets the following specifications to guarantee a smooth development experience:
+
+- [Hardware Requirements](../../guides/02-installation/06-hardware.md)
+
+### System Requirements
+
+Make sure your operating system meets the specific requirements for Jan development:
+
+- [Windows](../../install/windows/#system-requirements)
+- [MacOS](../../install/mac/#system-requirements)
+- [Linux](../../install/linux/#system-requirements)
+
+## Prerequisites
+
+- [Node.js](https://nodejs.org/en/) (version 20.0.0 or higher)
+- [yarn](https://yarnpkg.com/) (version 1.22.0 or higher)
+- [make](https://www.gnu.org/software/make/) (version 3.81 or higher)
+
+## Instructions
+
+1. **Clone the Repository:**
+
+```bash
+git clone https://github.com/janhq/jan
+cd jan
+git checkout -b DESIRED_BRANCH
+```
+
+2. **Install Dependencies**
+
+```bash
+yarn install
+```
+
+3. **Run Development and Use Jan Desktop**
+
+```bash
+make dev
+```
+
+This command starts the development server and opens the Jan Desktop app.
+
+## For Production Build
+
+```bash
+# Do steps 1 and 2 in the previous section
+# Build the app
+make build
+```
+
+This will build the app MacOS (M1/M2/M3) for production (with code signing already done) and place the result in `/electron/dist` folder.
+
+## Troubleshooting
+
+If you run into any issues due to a broken build, please check the [Stuck on a Broken Build](../../troubleshooting/stuck-on-broken-build) guide.

docs/docs/guides/02-installation/01-mac.md

@@ -12,11 +12,16 @@ keywords:
     conversational AI,
     no-subscription fee,
     large language model,
+    installation guide,
   ]
 ---
 
 # Installing Jan on MacOS
 
+## System Requirements
+
+Ensure that your MacOS version is 13 or higher to run Jan.
+
 ## Installation
 
 Jan is available for download via our homepage, [https://jan.ai/](https://jan.ai/).

docs/docs/guides/02-installation/02-windows.md

@@ -12,11 +12,23 @@ keywords:
     conversational AI,
     no-subscription fee,
     large language model,
+    installation guide,
   ]
 ---
 
 # Installing Jan on Windows
 
+## System Requirements
+
+Ensure that your system meets the following requirements:
+
+- Windows 10 or higher is required to run Jan.
+
+To enable GPU support, you will need:
+
+- NVIDIA GPU with CUDA Toolkit 11.7 or higher
+- NVIDIA driver 470.63.01 or higher
+
 ## Installation
 
 Jan is available for download via our homepage, [https://jan.ai](https://jan.ai/).
@@ -59,13 +71,3 @@ To remove all user data associated with Jan, you can delete the `/jan` directory
 cd C:\Users\%USERNAME%\AppData\Roaming
 rmdir /S jan
 ```
-
-## Troubleshooting
-
-### Microsoft Defender
-
-**Error: "Microsoft Defender SmartScreen prevented an unrecognized app from starting"**
-
-Windows Defender may display the above warning when running the Jan Installer, as a standard security measure.
-
-To proceed, select the "More info" option and select the "Run Anyway" option to continue with the installation.

docs/docs/guides/02-installation/03-linux.md

@@ -12,11 +12,24 @@ keywords:
     conversational AI,
     no-subscription fee,
     large language model,
+    installation guide,
   ]
 ---
 
 # Installing Jan on Linux
 
+## System Requirements
+
+Ensure that your system meets the following requirements:
+
+- glibc 2.27 or higher (check with `ldd --version`)
+- gcc 11, g++ 11, cpp 11, or higher, refer to this [link](https://jan.ai/guides/troubleshooting/gpu-not-used/#specific-requirements-for-linux) for more information.
+
+To enable GPU support, you will need:
+
+- NVIDIA GPU with CUDA Toolkit 11.7 or higher
+- NVIDIA driver 470.63.01 or higher
+
 ## Installation
 
 Jan is available for download via our homepage, [https://jan.ai](https://jan.ai/).
@@ -66,7 +79,6 @@ jan-linux-amd64-{version}.deb
 # AppImage
 jan-linux-x86_64-{version}.AppImage
 ```
-```
 
 ## Uninstall Jan
 

docs/docs/guides/04-using-models/03-integrate-with-remote-server.mdx

@@ -65,6 +65,13 @@ Navigate to the `~/jan/models` folder. Create a folder named `gpt-3.5-turbo-16k`
 }
 ```
 
+:::tip
+
+- You can find the list of available models in the [OpenAI Platform](https://platform.openai.com/docs/models/overview).
+- Please note that the `id` property need to match the model name in the list. For example, if you want to use the [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo), you need to set the `id` property as `gpt-4-1106-preview`.
+
+:::
+
 ### 2. Configure OpenAI API Keys
 
 You can find your API keys in the [OpenAI Platform](https://platform.openai.com/api-keys) and set the OpenAI API keys in `~/jan/engines/openai.json` file.

docs/docs/guides/08-troubleshooting/02-somethings-amiss.mdx

@@ -45,7 +45,9 @@ This may occur due to several reasons. Please follow these steps to resolve it:
 
 5. If you are on Nvidia GPUs, please download [Cuda](https://developer.nvidia.com/cuda-downloads).
 
-6. When [checking app logs](https://jan.ai/troubleshooting/how-to-get-error-logs/), if you encounter the error log `Bind address failed at 127.0.0.1:3928`, it indicates that the port used by Nitro might already be in use. Use the following commands to check the port status:
+6. If you're using Linux, please ensure that your system meets the following requirements gcc 11, g++ 11, cpp 11, or higher, refer to this [link](https://jan.ai/guides/troubleshooting/gpu-not-used/#specific-requirements-for-linux) for more information.
+
+7. When [checking app logs](https://jan.ai/troubleshooting/how-to-get-error-logs/), if you encounter the error log `Bind address failed at 127.0.0.1:3928`, it indicates that the port used by Nitro might already be in use. Use the following commands to check the port status:
 
 <Tabs groupId="operating-systems">
   <TabItem value="mac" label="macOS">

docs/docs/guides/08-troubleshooting/06-unexpected-token.mdx

@@ -17,4 +17,8 @@ keywords:
   ]
 ---
 
-1. You may receive an error response `Error occurred: Unexpected token '<', "<!DOCTYPE"...is not valid JSON`, when you start a chat with OpenAI models. Using a VPN may help fix the issue.
+You may receive an error response `Error occurred: Unexpected token '<', "<!DOCTYPE"...is not valid JSON`, when you start a chat with OpenAI models. 
+
+1. Check that you added an OpenAI API key. You can get an API key from OpenAI's [developer platform](https://platform.openai.com/). Alternatively, we recommend you download a local model from Jan Hub, which remains free to use and runs on your own computer!
+
+2. Using a VPN may help fix the issue.

docs/docs/guides/08-troubleshooting/07-undefined-issue.mdx

@@ -0,0 +1,26 @@
+---
+title: Undefined Issue
+slug: /troubleshooting/undefined-issue
+description: Undefined issue troubleshooting guide.
+keywords:
+  [
+    Jan AI,
+    Jan,
+    ChatGPT alternative,
+    local AI,
+    private AI,
+    conversational AI,
+    no-subscription fee,
+    large language model,
+    troubleshooting,
+    undefined issue,
+  ]
+---
+
+You may encounter an "undefined" issue when using Jan. Here are some troubleshooting steps to help you resolve the issue.
+
+1. Try wiping the Jan folder and reopening the Jan app and see if the issue persists.
+2. If the issue persists, try to go `~/jan/extensions/@janhq/inference-nitro-extensions/dist/bin/<your-os>/nitro` and run the nitro manually and see if you get any error messages.
+3. Resolve the error messages you get from the nitro and see if the issue persists.
+4. Reopen the Jan app and see if the issue is resolved.
+5. If the issue persists, please share with us the [app logs](https://jan.ai/troubleshooting/how-to-get-error-logs/) via [Jan Discord](https://discord.gg/mY69SZaMaC).

docs/docs/template/QA_script.md

@@ -1,6 +1,6 @@
 # [Release Version] QA Script 
 
-**Release Version:**
+**Release Version:** v0.4.6
 
 **Operating System:**
 
@@ -25,10 +25,10 @@
 
 ### 3. Users uninstall app
 
-- [ ] :key: Check that the uninstallation process removes all components of the app from the system.
+- [ ] :key::warning: Check that the uninstallation process removes the app successfully from the system.
 - [ ] Clean the Jan root directory and open the app to check if it creates all the necessary folders, especially models and extensions.
 - [ ] When updating the app, check if the `/models` directory has any JSON files that change according to the update.
-- [ ] Verify if updating the app also updates extensions correctly (test functionality changes; support notifications for necessary tests with each version related to extensions update).
+- [ ] Verify if updating the app also updates extensions correctly (test functionality changes, support notifications for necessary tests with each version related to extensions update).
 
 ### 4. Users close app
 
@@ -60,49 +60,45 @@
 - [ ] :key: Ensure that the conversation thread is maintained without any loss of data upon sending multiple messages.
 - [ ] Test for the ability to send different types of messages (e.g., text, emojis, code blocks).
 - [ ] :key: Validate the scroll functionality in the chat window for lengthy conversations.
-- [ ] Check if the user can renew responses multiple times.
 - [ ] Check if the user can copy the response.
 - [ ] Check if the user can delete responses.
-- [ ] :warning: Test if the user deletes the message midway, then the assistant stops that response.
 - [ ] :key: Check the `clear message` button works.
 - [ ] :key: Check the `delete entire chat` works.
-- [ ] :warning: Check if deleting all the chat retains the system prompt.
+- [ ] Check if deleting all the chat retains the system prompt.
 - [ ] Check the output format of the AI (code blocks, JSON, markdown, ...).
 - [ ] :key: Validate that there is appropriate error handling and messaging if the assistant fails to respond.
 - [ ] Test assistant's ability to maintain context over multiple exchanges.
 - [ ] :key: Check the `create new chat` button works correctly
 - [ ] Confirm that by changing `models` mid-thread the app can still handle it.
-- [ ] Check that by changing `instructions` mid-thread the app can still handle it.
+- [ ] Check the `regenerate` button renews the response (single / multiple times).
-- [ ] Check the `regenerate` button renews the response.
+- [ ] Check the `Instructions` update correctly after the user updates it midway (mid-thread).
-- [ ] Check the `Instructions` update correctly after the user updates it midway.
 
 ### 2. Users can customize chat settings like model parameters via both the GUI & thread.json
 
-- [ ] :key: Confirm that the chat settings options are accessible via the GUI.
+- [ ] :key: Confirm that the Threads settings options are accessible.
 - [ ] Test the functionality to adjust model parameters (e.g., Temperature, Top K, Top P) from the GUI and verify they are reflected in the chat behavior.
 - [ ] :key: Ensure that changes can be saved and persisted between sessions.
 - [ ] Validate that users can access and modify the thread.json file.
 - [ ] :key: Check that changes made in thread.json are correctly applied to the chat session upon reload or restart.
-- [ ] Verify if there is a revert option to go back to previous settings after changes are made.
-- [ ] Test for user feedback or confirmation after saving changes to settings.
 - [ ] Check the maximum and minimum limits of the adjustable parameters and how they affect the assistant's responses.
 - [ ] :key: Validate user permissions for those who can change settings and persist them.
 - [ ] :key: Ensure that users switch between threads with different models, the app can handle it.
 
-### 3. Users can click on a history thread
+### 3. Model dropdown
+- [ ] :key: Model list should highlight recommended based on user RAM
+- [ ] Model size should display (for both installed and imported models)
 
+### 4. Users can click on a history thread
 - [ ] Test the ability to click on any thread in the history panel.
 - [ ] :key: Verify that clicking a thread brings up the past conversation in the main chat window.
 - [ ] :key: Ensure that the selected thread is highlighted or otherwise indicated in the history panel.
 - [ ] Confirm that the chat window displays the entire conversation from the selected history thread without any missing messages.
 - [ ] :key: Check the performance and accuracy of the history feature when dealing with a large number of threads.
 - [ ] Validate that historical threads reflect the exact state of the chat at that time, including settings.
-- [ ] :key: :warning: Test the search functionality within the history panel for quick navigation.
 - [ ] :key: Verify the ability to delete or clean old threads.
 - [ ] :key: Confirm that changing the title of the thread updates correctly.
 
-### 4. Users can config instructions for the assistant.
+### 5. Users can config instructions for the assistant.
-
 - [ ] Ensure there is a clear interface to input or change instructions for the assistant.
 - [ ] Test if the instructions set by the user are being followed by the assistant in subsequent conversations.
 - [ ] :key: Validate that changes to instructions are updated in real time and do not require a restart of the application or session.
@@ -112,6 +108,8 @@
 - [ ] Validate that instructions can be saved with descriptive names for easy retrieval.
 - [ ] :key: Check if the assistant can handle conflicting instructions and how it resolves them.
 - [ ] Ensure that instruction configurations are documented for user reference.
+- [ ] :key: RAG - Users can import documents and the system should process queries about the uploaded file, providing accurate and appropriate responses in the conversation thread.
+
 
 ## D. Hub
 
@@ -125,8 +123,7 @@
 
 - [ ] Display the best model for their RAM at the top.
 - [ ] :key: Ensure that models are labeled with RAM requirements and compatibility.
-- [ ] :key: Validate that the download function is disabled for models that exceed the user's system capabilities.
+- [ ] :warning: Test that the platform provides alternative recommendations for models not suitable due to RAM limitations.
-- [ ] Test that the platform provides alternative recommendations for models not suitable due to RAM limitations.
 - [ ] :key: Check the download model functionality and validate if the cancel download feature works correctly.
 
 ### 3. Users can download models via a HuggingFace URL (coming soon)
@@ -139,7 +136,7 @@
 
 - [ ] :key: Have clear instructions so users can do their own.
 - [ ] :key: Ensure the new model updates after restarting the app.
-- [ ] Ensure it raises clear errors for users to fix the problem while adding a new model.
+- [ ] :warning:Ensure it raises clear errors for users to fix the problem while adding a new model.
 
 ### 5. Users can use the model as they want
 
@@ -149,9 +146,13 @@
 - [ ] Check if starting another model stops the other model entirely.
 - [ ] Check the `Explore models` navigate correctly to the model panel.
 - [ ] :key: Check when deleting a model it will delete all the files on the user's computer.
-- [ ] The recommended tags should present right for the user's hardware.
+- [ ] :warning:The recommended tags should present right for the user's hardware.
 - [ ] Assess that the descriptions of models are accurate and informative.
 
+### 6. Users can Integrate With a Remote Server
+- [ ] :key: Import openAI GPT model https://jan.ai/guides/using-models/integrate-with-remote-server/ and the model displayed in Hub / Thread dropdown
+- [ ] Users can use the remote model properly
+
 ## E. System Monitor
 
 ### 1. Users can see disk and RAM utilization
@@ -181,7 +182,7 @@
 - [ ] Confirm that the application saves the theme preference and persists it across sessions.
 - [ ] Validate that all elements of the UI are compatible with the theme changes and maintain legibility and contrast.
 
-### 2. Users change the extensions
+### 2. Users change the extensions [TBU]
 
 - [ ] Confirm that the `Extensions` tab lists all available plugins.
 - [ ] :key: Test the toggle switch for each plugin to ensure it enables or disables the plugin correctly.
@@ -208,3 +209,19 @@
 - [ ] :key: Test that the application prevents the installation of incompatible or corrupt plugin files.
 - [ ] :key: Check that the user can uninstall or disable custom plugins as easily as pre-installed ones.
 - [ ] Verify that the application's performance remains stable after the installation of custom plugins.
+
+### 5. Advanced Settings
+- [ ] Attemp to test downloading model from hub using **HTTP Proxy** [guideline](https://github.com/janhq/jan/pull/1562)
+- [ ] Users can move **Jan data folder**
+- [ ] Users can click on Reset button to **factory reset** app settings to its original state & delete all usage data.
+
+## G. Local API server
+
+### 1. Local Server Usage with Server Options
+- [ ] :key: Explore API Reference: Swagger API for sending/receiving requests
+    - [ ] Use default server option
+    - [ ] Configure and use custom server options
+- [ ] Test starting/stopping the local API server with different Model/Model settings
+- [ ] Server logs captured with correct Server Options provided
+- [ ] Verify functionality of Open logs/Clear feature
+- [ ] Ensure that threads and other functions impacting the model are disabled while the local server is running

docs/openapi/jan.yaml

@@ -67,20 +67,31 @@ paths:
       x-codeSamples:
         - lang: cURL
           source: |
-            curl http://localhost:1337/v1/chat/completions \
+            curl -X 'POST' \
-              -H "Content-Type: application/json" \
+              'http://localhost:1337/v1/chat/completions' \
+              -H 'accept: application/json' \
+              -H 'Content-Type: application/json' \
               -d '{
-                "model": "tinyllama-1.1b",
                 "messages": [
                   {
-                    "role": "system",
+                    "content": "You are a helpful assistant.",
-                    "content": "You are a helpful assistant."
+                    "role": "system"
                   },
                   {
-                    "role": "user",
+                    "content": "Hello!",
-                    "content": "Hello!"
+                    "role": "user"
                   }
-                ]
+                ],
+                "model": "tinyllama-1.1b",
+                "stream": true,
+                "max_tokens": 2048,
+                "stop": [
+                  "hello"
+                ],
+                "frequency_penalty": 0,
+                "presence_penalty": 0,
+                "temperature": 0.7,
+                "top_p": 0.95
               }'
   /models:
     get:
@@ -103,7 +114,9 @@ paths:
       x-codeSamples:
         - lang: cURL
           source: |
-            curl http://localhost:1337/v1/models
+            curl -X 'GET' \
+              'http://localhost:1337/v1/models' \
+              -H 'accept: application/json'
   "/models/download/{model_id}":
     get:
       operationId: downloadModel
@@ -131,7 +144,9 @@ paths:
       x-codeSamples:
         - lang: cURL
           source: |
-            curl -X POST http://localhost:1337/v1/models/download/{model_id}
+            curl -X 'GET' \
+              'http://localhost:1337/v1/models/download/{model_id}' \
+              -H 'accept: application/json'
   "/models/{model_id}":
     get:
       operationId: retrieveModel
@@ -162,7 +177,9 @@ paths:
       x-codeSamples:
         - lang: cURL
           source: |
-            curl http://localhost:1337/v1/models/{model_id}
+            curl -X 'GET' \
+              'http://localhost:1337/v1/models/{model_id}' \
+              -H 'accept: application/json'
     delete:
       operationId: deleteModel
       tags:
@@ -191,7 +208,9 @@ paths:
       x-codeSamples:
         - lang: cURL
           source: |
-            curl -X DELETE http://localhost:1337/v1/models/{model_id}
+            curl -X 'DELETE' \
+              'http://localhost:1337/v1/models/{model_id}' \
+              -H 'accept: application/json'
   /threads:
     post:
       operationId: createThread

docs/openapi/specs/messages.yaml

@@ -1,3 +1,4 @@
+---
 components:
   schemas:
     MessageObject:
@@ -75,7 +76,7 @@ components:
           example: msg_abc123
         object:
           type: string
-          description: "Type of the object, indicating it's a thread message."
+          description: Type of the object, indicating it's a thread message.
           default: thread.message
         created_at:
           type: integer
@@ -88,7 +89,7 @@ components:
           example: thread_abc123
         role:
           type: string
-          description: "Role of the sender, either 'user' or 'assistant'."
+          description: Role of the sender, either 'user' or 'assistant'.
           example: user
         content:
           type: array
@@ -97,7 +98,7 @@ components:
             properties:
               type:
                 type: string
-                description: "Type of content, e.g., 'text'."
+                description: Type of content, e.g., 'text'.
                 example: text
               text:
                 type: object
@@ -110,21 +111,21 @@ components:
                     type: array
                     items:
                       type: string
-                    description: "Annotations for the text content, if any."
+                    description: Annotations for the text content, if any.
                     example: []
         file_ids:
           type: array
           items:
             type: string
-          description: "Array of file IDs associated with the message, if any."
+          description: Array of file IDs associated with the message, if any.
           example: []
         assistant_id:
           type: string
-          description: "Identifier of the assistant involved in the message, if applicable."
+          description: Identifier of the assistant involved in the message, if applicable.
           example: null
         run_id:
           type: string
-          description: "Run ID associated with the message, if applicable."
+          description: Run ID associated with the message, if applicable.
           example: null
         metadata:
           type: object
@@ -139,7 +140,7 @@ components:
           example: msg_abc123
         object:
           type: string
-          description: "Type of the object, indicating it's a thread message."
+          description: Type of the object, indicating it's a thread message.
           example: thread.message
         created_at:
           type: integer
@@ -152,7 +153,7 @@ components:
           example: thread_abc123
         role:
           type: string
-          description: "Role of the sender, either 'user' or 'assistant'."
+          description: Role of the sender, either 'user' or 'assistant'.
           example: user
         content:
           type: array
@@ -161,7 +162,7 @@ components:
             properties:
               type:
                 type: string
-                description: "Type of content, e.g., 'text'."
+                description: Type of content, e.g., 'text'.
                 example: text
               text:
                 type: object
@@ -174,21 +175,21 @@ components:
                     type: array
                     items:
                       type: string
-                    description: "Annotations for the text content, if any."
+                    description: Annotations for the text content, if any.
                     example: []
         file_ids:
           type: array
           items:
             type: string
-          description: "Array of file IDs associated with the message, if any."
+          description: Array of file IDs associated with the message, if any.
           example: []
         assistant_id:
           type: string
-          description: "Identifier of the assistant involved in the message, if applicable."
+          description: Identifier of the assistant involved in the message, if applicable.
           example: null
         run_id:
           type: string
-          description: "Run ID associated with the message, if applicable."
+          description: Run ID associated with the message, if applicable.
           example: null
         metadata:
           type: object
@@ -199,7 +200,7 @@ components:
       properties:
         object:
           type: string
-          description: "Type of the object, indicating it's a list."
+          description: Type of the object, indicating it's a list.
           default: list
         data:
           type: array
@@ -226,7 +227,7 @@ components:
           example: msg_abc123
         object:
           type: string
-          description: "Type of the object, indicating it's a thread message."
+          description: Type of the object, indicating it's a thread message.
           example: thread.message
         created_at:
           type: integer
@@ -239,7 +240,7 @@ components:
           example: thread_abc123
         role:
           type: string
-          description: "Role of the sender, either 'user' or 'assistant'."
+          description: Role of the sender, either 'user' or 'assistant'.
           example: user
         content:
           type: array
@@ -248,7 +249,7 @@ components:
             properties:
               type:
                 type: string
-                description: "Type of content, e.g., 'text'."
+                description: Type of content, e.g., 'text'.
               text:
                 type: object
                 properties:
@@ -260,20 +261,20 @@ components:
                     type: array
                     items:
                       type: string
-                    description: "Annotations for the text content, if any."
+                    description: Annotations for the text content, if any.
         file_ids:
           type: array
           items:
             type: string
-          description: "Array of file IDs associated with the message, if any."
+          description: Array of file IDs associated with the message, if any.
           example: []
         assistant_id:
           type: string
-          description: "Identifier of the assistant involved in the message, if applicable."
+          description: Identifier of the assistant involved in the message, if applicable.
           example: null
         run_id:
           type: string
-          description: "Run ID associated with the message, if applicable."
+          description: Run ID associated with the message, if applicable.
           example: null
         metadata:
           type: object

docs/openapi/specs/models.yaml

@@ -18,114 +18,82 @@ components:
     Model:
       type: object
       properties:
-        type:
+        source_url:
           type: string
-          default: model
+          format: uri
-          description: The type of the object.
+          description: URL to the source of the model.
-        version:
+          example: https://huggingface.co/janhq/trinity-v1.2-GGUF/resolve/main/trinity-v1.2.Q4_K_M.gguf
-          type: string
-          default: "1"
-          description: The version number of the model.
         id:
           type: string
-          description: Unique identifier used in chat-completions model_name, matches
+          description:
+            Unique identifier used in chat-completions model_name, matches
             folder name.
-          example: zephyr-7b
+          example: trinity-v1.2-7b
+        object:
+          type: string
+          example: model
         name:
           type: string
           description: Name of the model.
-          example: Zephyr 7B
+          example: Trinity-v1.2 7B Q4
-        owned_by:
+        version:
           type: string
-          description: Compatibility field for OpenAI.
+          default: "1.0"
-          default: ""
+          description: The version number of the model.
-        created:
-          type: integer
-          format: int64
-          description: Unix timestamp representing the creation time.
         description:
           type: string
           description: Description of the model.
-        state:
+          example:
-          type: string
+            Trinity is an experimental model merge using the Slerp method.
-          enum:
+            Recommended for daily assistance purposes.
-            - null
-            - downloading
-            - ready
-            - starting
-            - stopping
-          description: Current state of the model.
         format:
           type: string
           description: State format of the model, distinct from the engine.
-          example: ggufv3
+          example: gguf
-        source:
-          type: array
-          items:
-            type: object
-            properties:
-              url:
-                format: uri
-                description: URL to the source of the model.
-                example: https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/blob/main/zephyr-7b-beta.Q4_K_M.gguf
-              filename:
-                type: string
-                description: Filename of the model.
-                example: zephyr-7b-beta.Q4_K_M.gguf
         settings:
           type: object
           properties:
             ctx_len:
-              type: string
+              type: integer
               description: Context length.
-              example: "4096"
+              example: 4096
-            ngl:
+            prompt_template:
               type: string
-              description: Number of layers.
+              example: "<|im_start|>system\n{system_message}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant"
-              example: "100"
-            embedding:
-              type: string
-              description: Indicates if embedding is enabled.
-              example: "true"
-            n_parallel:
-              type: string
-              description: Number of parallel processes.
-              example: "4"
           additionalProperties: false
         parameters:
           type: object
           properties:
             temperature:
-              type: string
+              example: 0.7
-              description: Temperature setting for the model.
-              example: "0.7"
-            token_limit:
-              type: string
-              description: Token limit for the model.
-              example: "4096"
-            top_k:
-              type: string
-              description: Top-k setting for the model.
-              example: "0"
             top_p:
-              type: string
+              example: 0.95
-              description: Top-p setting for the model.
-              example: "1"
             stream:
-              type: string
+              example: true
-              description: Indicates if streaming is enabled.
+            max_tokens:
-              example: "true"
+              example: 4096
+            stop:
+              example: []
+            frequency_penalty:
+              example: 0
+            presence_penalty:
+              example: 0
           additionalProperties: false
         metadata:
-          type: object
+          author:
-          description: Additional metadata.
-        assets:
-          type: array
-          items:
             type: string
-          description: List of assets related to the model.
+            example: Jan
-      required:
+          tags:
-        - source
+            example:
+              - 7B
+              - Merged
+              - Featured
+          size:
+            example: 4370000000,
+          cover:
+            example: https://raw.githubusercontent.com/janhq/jan/main/models/trinity-v1.2-7b/cover.png
+        engine:
+          example: nitro
     ModelObject:
       type: object
       properties:
@@ -133,7 +101,7 @@ components:
           type: string
           description: |
             The identifier of the model.
-          example: zephyr-7b
+          example: trinity-v1.2-7b
         object:
           type: string
           description: |
@@ -153,197 +121,89 @@ components:
     GetModelResponse:
       type: object
       properties:
+        source_url:
+          type: string
+          format: uri
+          description: URL to the source of the model.
+          example: https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf
         id:
           type: string
-          description: The identifier of the model.
+          description:
-          example: zephyr-7b
+            Unique identifier used in chat-completions model_name, matches
+            folder name.
+          example: mistral-ins-7b-q4
         object:
           type: string
-          description: Type of the object, indicating it's a model.
+          example: model
-          default: model
+        name:
-        created:
-          type: integer
-          format: int64
-          description: Unix timestamp representing the creation time of the model.
-        owned_by:
           type: string
-          description: The entity that owns the model.
+          description: Name of the model.
-          example: _
+          example: Mistral Instruct 7B Q4
-        state:
+        version:
           type: string
-          enum:
+          default: "1.0"
-            - not_downloaded
+          description: The version number of the model.
-            - downloaded
+        description:
-            - running
+          type: string
-            - stopped
+          description: Description of the model.
-          description: The current state of the model.
+          example:
-        source:
+            Trinity is an experimental model merge using the Slerp method.
-          type: array
+            Recommended for daily assistance purposes.
-          items:
+        format:
-            type: object
+          type: string
-            properties:
+          description: State format of the model, distinct from the engine.
-              url:
+          example: gguf
-                format: uri
+        settings:
-                description: URL to the source of the model.
-                example: https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/blob/main/zephyr-7b-beta.Q4_K_M.gguf
-              filename:
-                type: string
-                description: Filename of the model.
-                example: zephyr-7b-beta.Q4_K_M.gguf
-        engine_parameters:
-          type: object
-          properties:
-            pre_prompt:
-              type: string
-              description: Predefined prompt used for setting up internal configurations.
-              default: ""
-              example: Initial setup complete.
-            system_prompt:
-              type: string
-              description: Prefix used for system-level prompts.
-              default: "SYSTEM: "
-            user_prompt:
-              type: string
-              description: Prefix used for user prompts.
-              default: "USER: "
-            ai_prompt:
-              type: string
-              description: Prefix used for assistant prompts.
-              default: "ASSISTANT: "
-            ngl:
-              type: integer
-              description: Number of neural network layers loaded onto the GPU for
-                acceleration.
-              minimum: 0
-              maximum: 100
-              default: 100
-              example: 100
-            ctx_len:
-              type: integer
-              description: Context length for model operations, varies based on the specific
-                model.
-              minimum: 128
-              maximum: 4096
-              default: 4096
-              example: 4096
-            n_parallel:
-              type: integer
-              description: Number of parallel operations, relevant when continuous batching is
-                enabled.
-              minimum: 1
-              maximum: 10
-              default: 1
-              example: 4
-            cont_batching:
-              type: boolean
-              description: Indicates if continuous batching is used for processing.
-              default: false
-              example: false
-            cpu_threads:
-              type: integer
-              description: Number of threads allocated for CPU-based inference.
-              minimum: 1
-              example: 8
-            embedding:
-              type: boolean
-              description: Indicates if embedding layers are enabled in the model.
-              default: true
-              example: true
-        model_parameters:
           type: object
           properties:
             ctx_len:
               type: integer
-              description: Maximum context length the model can handle.
+              description: Context length.
-              minimum: 0
-              maximum: 4096
-              default: 4096
               example: 4096
-            ngl:
+            prompt_template:
-              type: integer
+              type: string
-              description: Number of layers in the neural network.
+              example: "[INST] {prompt} [/INST]"
-              minimum: 1
+          additionalProperties: false
-              maximum: 100
+        parameters:
-              default: 100
+          type: object
-              example: 100
+          properties:
-            embedding:
-              type: boolean
-              description: Indicates if embedding layers are used.
-              default: true
-              example: true
-            n_parallel:
-              type: integer
-              description: Number of parallel processes the model can run.
-              minimum: 1
-              maximum: 10
-              default: 1
-              example: 4
             temperature:
-              type: number
-              description: Controls randomness in model's responses. Higher values lead to
-                more random responses.
-              minimum: 0
-              maximum: 2
-              default: 0.7
               example: 0.7
-            token_limit:
-              type: integer
-              description: Maximum number of tokens the model can generate in a single
-                response.
-              minimum: 1
-              maximum: 4096
-              default: 4096
-              example: 4096
-            top_k:
-              type: integer
-              description: Limits the model to consider only the top k most likely next tokens
-                at each step.
-              minimum: 0
-              maximum: 100
-              default: 0
-              example: 0
             top_p:
-              type: number
+              example: 0.95
-              description: Nucleus sampling parameter. The model considers the smallest set of
+            stream:
-                tokens whose cumulative probability exceeds the top_p value.
+              example: true
-              minimum: 0
+            max_tokens:
-              maximum: 1
+              example: 4096
-              default: 1
+            stop:
-              example: 1
+              example: []
+            frequency_penalty:
+              example: 0
+            presence_penalty:
+              example: 0
+          additionalProperties: false
         metadata:
-          type: object
+          author:
-          properties:
+            type: string
-            engine:
+            example: MistralAI
-              type: string
+          tags:
-              description: The engine used by the model.
+            example:
-              enum:
+              - 7B
-                - nitro
+              - Featured
-                - openai
+              - Foundation Model
-                - hf_inference
+          size:
-            quantization:
+            example: 4370000000,
-              type: string
+          cover:
-              description: Quantization parameter of the model.
+            example: https://raw.githubusercontent.com/janhq/jan/main/models/mistral-ins-7b-q4/cover.png
-              example: Q3_K_L
+        engine:
-            size:
+          example: nitro
-              type: string
-              description: Size of the model.
-              example: 7B
-      required:
-        - id
-        - object
-        - created
-        - owned_by
-        - state
-        - source
-        - parameters
-        - metadata
     DeleteModelResponse:
       type: object
       properties:
         id:
           type: string
           description: The identifier of the model that was deleted.
-          example: model-zephyr-7B
+          example: mistral-ins-7b-q4
         object:
           type: string
           description: Type of the object, indicating it's a model.

docs/openapi/specs/threads.yaml

@@ -142,7 +142,7 @@ components:
                 example: Jan
               instructions:
                 type: string
-                description: |
+                description: >
                   The instruction of assistant, defaults to "Be my grammar corrector"
               model:
                 type: object

electron/handlers/download.ts

@@ -5,7 +5,11 @@ import request from 'request'
 import { createWriteStream, renameSync } from 'fs'
 import { DownloadEvent, DownloadRoute } from '@janhq/core'
 const progress = require('request-progress')
-import { DownloadManager, getJanDataFolderPath, normalizeFilePath } from '@janhq/core/node'
+import {
+  DownloadManager,
+  getJanDataFolderPath,
+  normalizeFilePath,
+} from '@janhq/core/node'
 
 export function handleDownloaderIPCs() {
   /**
@@ -56,20 +60,23 @@ export function handleDownloaderIPCs() {
    */
   ipcMain.handle(
     DownloadRoute.downloadFile,
-    async (_event, url, fileName, network) => {
+    async (_event, url, localPath, network) => {
       const strictSSL = !network?.ignoreSSL
       const proxy = network?.proxy?.startsWith('http')
         ? network.proxy
         : undefined
-
+      if (typeof localPath === 'string') {
-      if (typeof fileName === 'string') {
+        localPath = normalizeFilePath(localPath)
-        fileName = normalizeFilePath(fileName)
       }
-      const destination = resolve(getJanDataFolderPath(), fileName)
+      const array = localPath.split('/')
+      const fileName = array.pop() ?? ''
+      const modelId = array.pop() ?? ''
+
+      const destination = resolve(getJanDataFolderPath(), localPath)
       const rq = request({ url, strictSSL, proxy })
 
       // Put request to download manager instance
-      DownloadManager.instance.setRequest(fileName, rq)
+      DownloadManager.instance.setRequest(localPath, rq)
 
       // Downloading file to a temp file first
       const downloadingTempFile = `${destination}.download`
@@ -81,6 +88,7 @@ export function handleDownloaderIPCs() {
             {
               ...state,
               fileName,
+              modelId,
             }
           )
         })
@@ -90,11 +98,12 @@ export function handleDownloaderIPCs() {
             {
               fileName,
               err,
+              modelId,
             }
           )
         })
         .on('end', function () {
-          if (DownloadManager.instance.networkRequests[fileName]) {
+          if (DownloadManager.instance.networkRequests[localPath]) {
             // Finished downloading, rename temp file to actual file
             renameSync(downloadingTempFile, destination)
 
@@ -102,14 +111,16 @@ export function handleDownloaderIPCs() {
               DownloadEvent.onFileDownloadSuccess,
               {
                 fileName,
+                modelId,
               }
             )
-            DownloadManager.instance.setRequest(fileName, undefined)
+            DownloadManager.instance.setRequest(localPath, undefined)
           } else {
             WindowManager?.instance.currentWindow?.webContents.send(
               DownloadEvent.onFileDownloadError,
               {
                 fileName,
+                modelId,
                 err: { message: 'aborted' },
               }
             )

electron/handlers/fileManager.ts

@@ -38,6 +38,7 @@ export function handleFileMangerIPCs() {
     getResourcePath()
   )
 
+  // Handles the 'getUserHomePath' IPC event. This event is triggered to get the user home path.
   ipcMain.handle(FileManagerRoute.getUserHomePath, async (_event) =>
     app.getPath('home')
   )

electron/handlers/fs.ts

@@ -1,8 +1,7 @@
 import { ipcMain } from 'electron'
 
 import { getJanDataFolderPath, normalizeFilePath } from '@janhq/core/node'
-import fs from 'fs'
+import { FileSystemRoute } from '@janhq/core'
-import { FileManagerRoute, FileSystemRoute } from '@janhq/core'
 import { join } from 'path'
 /**
  * Handles file system operations.

electron/merge-latest-ymls.js

@@ -0,0 +1,27 @@
+const yaml = require('js-yaml')
+const fs = require('fs')
+
+// get two file paths from arguments:
+const [, , ...args] = process.argv
+const file1 = args[0]
+const file2 = args[1]
+const file3 = args[2]
+
+// check that all arguments are present and throw error instead
+if (!file1 || !file2 || !file3) {
+  throw new Error('Please provide 3 file paths as arguments: path to file1, to file2 and destination path')
+}
+
+const doc1 = yaml.load(fs.readFileSync(file1, 'utf8'))
+console.log('doc1: ', doc1)
+
+const doc2 = yaml.load(fs.readFileSync(file2, 'utf8'))
+console.log('doc2: ', doc2)
+
+const merged = { ...doc1, ...doc2 }
+merged.files.push(...doc1.files)
+
+console.log('merged', merged)
+
+const mergedYml = yaml.dump(merged)
+fs.writeFileSync(file3, mergedYml, 'utf8')

electron/package.json

@@ -57,16 +57,17 @@
   "scripts": {
     "lint": "eslint . --ext \".js,.jsx,.ts,.tsx\"",
     "test:e2e": "playwright test --workers=1",
-    "dev": "tsc -p . && electron .",
+    "copy:assets": "rimraf --glob \"./pre-install/*.tgz\" && cpx \"../pre-install/*.tgz\" \"./pre-install\"",
-    "build": "run-script-os",
+    "dev": "yarn copy:assets && tsc -p . && electron .",
-    "build:test": "run-script-os",
+    "build": "yarn copy:assets && run-script-os",
+    "build:test": "yarn copy:assets && run-script-os",
     "build:test:darwin": "tsc -p . && electron-builder -p never -m --dir",
     "build:test:win32": "tsc -p . && electron-builder -p never -w --dir",
     "build:test:linux": "tsc -p . && electron-builder -p never -l --dir",
     "build:darwin": "tsc -p . && electron-builder -p never -m --x64 --arm64",
     "build:win32": "tsc -p . && electron-builder -p never -w",
     "build:linux": "tsc -p . && electron-builder -p never -l deb -l AppImage",
-    "build:publish": "run-script-os",
+    "build:publish": "yarn copy:assets && run-script-os",
     "build:publish:darwin": "tsc -p . && electron-builder -p always -m --x64 --arm64",
     "build:publish:win32": "tsc -p . && electron-builder -p always -w",
     "build:publish:linux": "tsc -p . && electron-builder -p always -l deb -l AppImage"

electron/playwright.config.ts

@@ -1,9 +1,16 @@
 import { PlaywrightTestConfig } from '@playwright/test'
 
 const config: PlaywrightTestConfig = {
-  testDir: './tests',
+  testDir: './tests/e2e',
   retries: 0,
   globalTimeout: 300000,
+  use: {
+    screenshot: 'only-on-failure',
+    video: 'retain-on-failure',
+    trace: 'retain-on-failure',
+  },
+
+  reporter: [['html', { outputFolder: './playwright-report' }]],
 }
 
 export default config

electron/tests/e2e/hub.e2e.spec.ts

@@ -0,0 +1,34 @@
+import {
+  page,
+  test,
+  setupElectron,
+  teardownElectron,
+  TIMEOUT,
+} from '../pages/basePage'
+import { expect } from '@playwright/test'
+
+test.beforeAll(async () => {
+  const appInfo = await setupElectron()
+  expect(appInfo.asar).toBe(true)
+  expect(appInfo.executable).toBeTruthy()
+  expect(appInfo.main).toBeTruthy()
+  expect(appInfo.name).toBe('jan')
+  expect(appInfo.packageJson).toBeTruthy()
+  expect(appInfo.packageJson.name).toBe('jan')
+  expect(appInfo.platform).toBeTruthy()
+  expect(appInfo.platform).toBe(process.platform)
+  expect(appInfo.resourcesDir).toBeTruthy()
+})
+
+test.afterAll(async () => {
+  await teardownElectron()
+})
+
+test('explores hub', async () => {
+  await page.getByTestId('Hub').first().click({
+    timeout: TIMEOUT,
+  })
+  await page.getByTestId('hub-container-test-id').isVisible({
+    timeout: TIMEOUT,
+  })
+})

electron/tests/e2e/navigation.e2e.spec.ts

@@ -0,0 +1,38 @@
+import { expect } from '@playwright/test'
+import {
+  page,
+  setupElectron,
+  TIMEOUT,
+  test,
+  teardownElectron,
+} from '../pages/basePage'
+
+test.beforeAll(async () => {
+  await setupElectron()
+})
+
+test.afterAll(async () => {
+  await teardownElectron()
+})
+
+test('renders left navigation panel', async () => {
+  const systemMonitorBtn = await page
+    .getByTestId('System Monitor')
+    .first()
+    .isEnabled({
+      timeout: TIMEOUT,
+    })
+  const settingsBtn = await page
+    .getByTestId('Thread')
+    .first()
+    .isEnabled({ timeout: TIMEOUT })
+  expect([systemMonitorBtn, settingsBtn].filter((e) => !e).length).toBe(0)
+  // Chat section should be there
+  await page.getByTestId('Local API Server').first().click({
+    timeout: TIMEOUT,
+  })
+  const localServer = page.getByTestId('local-server-testid').first()
+  await expect(localServer).toBeVisible({
+    timeout: TIMEOUT,
+  })
+})

electron/tests/e2e/settings.e2e.spec.ts

@@ -0,0 +1,23 @@
+import { expect } from '@playwright/test'
+
+import {
+  setupElectron,
+  teardownElectron,
+  test,
+  page,
+  TIMEOUT,
+} from '../pages/basePage'
+
+test.beforeAll(async () => {
+  await setupElectron()
+})
+
+test.afterAll(async () => {
+  await teardownElectron()
+})
+
+test('shows settings', async () => {
+  await page.getByTestId('Settings').first().click({ timeout: TIMEOUT })
+  const settingDescription = page.getByTestId('testid-setting-description')
+  await expect(settingDescription).toBeVisible({ timeout: TIMEOUT })
+})

electron/tests/hub.e2e.spec.ts

@@ -1,48 +0,0 @@
-import { _electron as electron } from 'playwright'
-import { ElectronApplication, Page, expect, test } from '@playwright/test'
-
-import {
-  findLatestBuild,
-  parseElectronApp,
-  stubDialog,
-} from 'electron-playwright-helpers'
-
-let electronApp: ElectronApplication
-let page: Page
-const TIMEOUT: number = parseInt(process.env.TEST_TIMEOUT || '300000')
-
-test.beforeAll(async () => {
-  process.env.CI = 'e2e'
-
-  const latestBuild = findLatestBuild('dist')
-  expect(latestBuild).toBeTruthy()
-
-  // parse the packaged Electron app and find paths and other info
-  const appInfo = parseElectronApp(latestBuild)
-  expect(appInfo).toBeTruthy()
-
-  electronApp = await electron.launch({
-    args: [appInfo.main], // main file from package.json
-    executablePath: appInfo.executable, // path to the Electron executable
-  })
-  await stubDialog(electronApp, 'showMessageBox', { response: 1 })
-
-  page = await electronApp.firstWindow({
-    timeout: TIMEOUT,
-  })
-})
-
-test.afterAll(async () => {
-  await electronApp.close()
-  await page.close()
-})
-
-test('explores hub', async () => {
-  test.setTimeout(TIMEOUT)
-  await page.getByTestId('Hub').first().click({
-    timeout: TIMEOUT,
-  })
-  await page.getByTestId('hub-container-test-id').isVisible({
-    timeout: TIMEOUT,
-  })
-})

electron/tests/navigation.e2e.spec.ts

@@ -1,61 +0,0 @@
-import { _electron as electron } from 'playwright'
-import { ElectronApplication, Page, expect, test } from '@playwright/test'
-
-import {
-  findLatestBuild,
-  parseElectronApp,
-  stubDialog,
-} from 'electron-playwright-helpers'
-
-let electronApp: ElectronApplication
-let page: Page
-const TIMEOUT: number = parseInt(process.env.TEST_TIMEOUT || '300000')
-
-test.beforeAll(async () => {
-  process.env.CI = 'e2e'
-
-  const latestBuild = findLatestBuild('dist')
-  expect(latestBuild).toBeTruthy()
-
-  // parse the packaged Electron app and find paths and other info
-  const appInfo = parseElectronApp(latestBuild)
-  expect(appInfo).toBeTruthy()
-
-  electronApp = await electron.launch({
-    args: [appInfo.main], // main file from package.json
-    executablePath: appInfo.executable, // path to the Electron executable
-  })
-  await stubDialog(electronApp, 'showMessageBox', { response: 1 })
-
-  page = await electronApp.firstWindow({
-    timeout: TIMEOUT,
-  })
-})
-
-test.afterAll(async () => {
-  await electronApp.close()
-  await page.close()
-})
-
-test('renders left navigation panel', async () => {
-  test.setTimeout(TIMEOUT)
-  const systemMonitorBtn = await page
-    .getByTestId('System Monitor')
-    .first()
-    .isEnabled({
-      timeout: TIMEOUT,
-    })
-  const settingsBtn = await page
-    .getByTestId('Thread')
-    .first()
-    .isEnabled({ timeout: TIMEOUT })
-  expect([systemMonitorBtn, settingsBtn].filter((e) => !e).length).toBe(0)
-  // Chat section should be there
-  await page.getByTestId('Local API Server').first().click({
-    timeout: TIMEOUT,
-  })
-  const localServer = await page.getByTestId('local-server-testid').first()
-  await expect(localServer).toBeVisible({
-    timeout: TIMEOUT,
-  })
-})

electron/tests/pages/basePage.ts

@@ -0,0 +1,67 @@
+import {
+  expect,
+  test as base,
+  _electron as electron,
+  ElectronApplication,
+  Page,
+} from '@playwright/test'
+import {
+  findLatestBuild,
+  parseElectronApp,
+  stubDialog,
+} from 'electron-playwright-helpers'
+
+export const TIMEOUT: number = parseInt(process.env.TEST_TIMEOUT || '300000')
+
+export let electronApp: ElectronApplication
+export let page: Page
+
+export async function setupElectron() {
+  process.env.CI = 'e2e'
+
+  const latestBuild = findLatestBuild('dist')
+  expect(latestBuild).toBeTruthy()
+
+  // parse the packaged Electron app and find paths and other info
+  const appInfo = parseElectronApp(latestBuild)
+  expect(appInfo).toBeTruthy()
+
+  electronApp = await electron.launch({
+    args: [appInfo.main], // main file from package.json
+    executablePath: appInfo.executable, // path to the Electron executable
+  })
+  await stubDialog(electronApp, 'showMessageBox', { response: 1 })
+
+  page = await electronApp.firstWindow({
+    timeout: TIMEOUT,
+  })
+  // Return appInfo for future use
+  return appInfo
+}
+
+export async function teardownElectron() {
+  await page.close()
+  await electronApp.close()
+}
+
+export const test = base.extend<{
+  attachScreenshotsToReport: void
+}>({
+  attachScreenshotsToReport: [
+    async ({ request }, use, testInfo) => {
+      await use()
+
+      // After the test, we can check whether the test passed or failed.
+      if (testInfo.status !== testInfo.expectedStatus) {
+        const screenshot = await page.screenshot()
+        await testInfo.attach('screenshot', {
+          body: screenshot,
+          contentType: 'image/png',
+        })
+      }
+    },
+    { auto: true },
+  ],
+})
+
+test.setTimeout(TIMEOUT)

electron/tests/settings.e2e.spec.ts

@@ -1,45 +0,0 @@
-import { _electron as electron } from 'playwright'
-import { ElectronApplication, Page, expect, test } from '@playwright/test'
-
-import {
-  findLatestBuild,
-  parseElectronApp,
-  stubDialog,
-} from 'electron-playwright-helpers'
-
-let electronApp: ElectronApplication
-let page: Page
-const TIMEOUT: number = parseInt(process.env.TEST_TIMEOUT || '300000')
-
-test.beforeAll(async () => {
-  process.env.CI = 'e2e'
-
-  const latestBuild = findLatestBuild('dist')
-  expect(latestBuild).toBeTruthy()
-
-  // parse the packaged Electron app and find paths and other info
-  const appInfo = parseElectronApp(latestBuild)
-  expect(appInfo).toBeTruthy()
-
-  electronApp = await electron.launch({
-    args: [appInfo.main], // main file from package.json
-    executablePath: appInfo.executable, // path to the Electron executable
-  })
-  await stubDialog(electronApp, 'showMessageBox', { response: 1 })
-
-  page = await electronApp.firstWindow({
-    timeout: TIMEOUT,
-  })
-})
-
-test.afterAll(async () => {
-  await electronApp.close()
-  await page.close()
-})
-
-test('shows settings', async () => {
-  test.setTimeout(TIMEOUT)
-  await page.getByTestId('Settings').first().click({ timeout: TIMEOUT })
-  const settingDescription = page.getByTestId('testid-setting-description')
-  await expect(settingDescription).toBeVisible({ timeout: TIMEOUT })
-})

extensions/assistant-extension/package.json

@@ -8,9 +8,9 @@
   "license": "AGPL-3.0",
   "scripts": {
     "build": "tsc --module commonjs && rollup -c rollup.config.ts",
-    "build:publish:linux": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install",
+    "build:publish:linux": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install",
-    "build:publish:darwin": "rimraf *.tgz --glob && npm run build && ../../.github/scripts/auto-sign.sh && npm pack && cpx *.tgz ../../electron/pre-install",
+    "build:publish:darwin": "rimraf *.tgz --glob && npm run build && ../../.github/scripts/auto-sign.sh && npm pack && cpx *.tgz ../../pre-install",
-    "build:publish:win32": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install",
+    "build:publish:win32": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install",
     "build:publish": "run-script-os"
   },
   "devDependencies": {

extensions/assistant-extension/src/index.ts

@@ -9,6 +9,7 @@ import {
   joinPath,
   executeOnMain,
   AssistantExtension,
+  AssistantEvent,
 } from "@janhq/core";
 
 export default class JanAssistantExtension extends AssistantExtension {
@@ -21,7 +22,7 @@ export default class JanAssistantExtension extends AssistantExtension {
   async onLoad() {
     // making the assistant directory
     const assistantDirExist = await fs.existsSync(
-      JanAssistantExtension._homeDir,
+      JanAssistantExtension._homeDir
     );
     if (
       localStorage.getItem(`${EXTENSION_NAME}-version`) !== VERSION ||
@@ -31,14 +32,16 @@ export default class JanAssistantExtension extends AssistantExtension {
         await fs.mkdirSync(JanAssistantExtension._homeDir);
 
       // Write assistant metadata
-      this.createJanAssistant();
+      await this.createJanAssistant();
       // Finished migration
       localStorage.setItem(`${EXTENSION_NAME}-version`, VERSION);
+      // Update the assistant list
+      events.emit(AssistantEvent.OnAssistantsUpdate, {});
     }
 
     // Events subscription
     events.on(MessageEvent.OnMessageSent, (data: MessageRequest) =>
-      JanAssistantExtension.handleMessageRequest(data, this),
+      JanAssistantExtension.handleMessageRequest(data, this)
     );
 
     events.on(InferenceEvent.OnInferenceStopped, () => {
@@ -53,7 +56,7 @@ export default class JanAssistantExtension extends AssistantExtension {
 
   private static async handleMessageRequest(
     data: MessageRequest,
-    instance: JanAssistantExtension,
+    instance: JanAssistantExtension
   ) {
     instance.isCancelled = false;
     instance.controller = new AbortController();
@@ -80,7 +83,7 @@ export default class JanAssistantExtension extends AssistantExtension {
           NODE,
           "toolRetrievalIngestNewDocument",
           docFile,
-          data.model?.proxyEngine,
+          data.model?.proxyEngine
         );
       }
     }
@@ -96,7 +99,7 @@ export default class JanAssistantExtension extends AssistantExtension {
         NODE,
         "toolRetrievalUpdateTextSplitter",
         data.thread.assistants[0].tools[0]?.settings?.chunk_size ?? 4000,
-        data.thread.assistants[0].tools[0]?.settings?.chunk_overlap ?? 200,
+        data.thread.assistants[0].tools[0]?.settings?.chunk_overlap ?? 200
       );
     }
 
@@ -110,7 +113,7 @@ export default class JanAssistantExtension extends AssistantExtension {
       const retrievalResult = await executeOnMain(
         NODE,
         "toolRetrievalQueryResult",
-        prompt,
+        prompt
       );
 
       // Update the message content
@@ -168,7 +171,7 @@ export default class JanAssistantExtension extends AssistantExtension {
     try {
       await fs.writeFileSync(
         assistantMetadataPath,
-        JSON.stringify(assistant, null, 2),
+        JSON.stringify(assistant, null, 2)
       );
     } catch (err) {
       console.error(err);
@@ -180,7 +183,7 @@ export default class JanAssistantExtension extends AssistantExtension {
     // get all the assistant metadata json
     const results: Assistant[] = [];
     const allFileName: string[] = await fs.readdirSync(
-      JanAssistantExtension._homeDir,
+      JanAssistantExtension._homeDir
     );
     for (const fileName of allFileName) {
       const filePath = await joinPath([
@@ -190,7 +193,7 @@ export default class JanAssistantExtension extends AssistantExtension {
 
       if (filePath.includes(".DS_Store")) continue;
       const jsonFiles: string[] = (await fs.readdirSync(filePath)).filter(
-        (file: string) => file === "assistant.json",
+        (file: string) => file === "assistant.json"
       );
 
       if (jsonFiles.length !== 1) {
@@ -200,7 +203,7 @@ export default class JanAssistantExtension extends AssistantExtension {
 
       const content = await fs.readFileSync(
         await joinPath([filePath, jsonFiles[0]]),
-        "utf-8",
+        "utf-8"
       );
       const assistant: Assistant =
         typeof content === "object" ? content : JSON.parse(content);

extensions/assistant-extension/src/node/tools/retrieval/index.ts

@@ -35,21 +35,19 @@ export class Retrieval {
     if (engine === "nitro") {
       this.embeddingModel = new OpenAIEmbeddings(
         { openAIApiKey: "nitro-embedding" },
-        { basePath: "http://127.0.0.1:3928/v1" }
+        { basePath: "http://127.0.0.1:3928/v1" },
       );
     } else {
       // Fallback to OpenAI Settings
       this.embeddingModel = new OpenAIEmbeddings({
-        configuration: {
+        openAIApiKey: settings.api_key,
-          apiKey: settings.api_key,
-        },
       });
     }
   }
 
   public ingestAgentKnowledge = async (
     filePath: string,
-    memoryPath: string
+    memoryPath: string,
   ): Promise<any> => {
     const loader = new PDFLoader(filePath, {
       splitPages: true,

extensions/conversational-extension/package.json

@@ -7,7 +7,7 @@
   "license": "MIT",
   "scripts": {
     "build": "tsc -b . && webpack --config webpack.config.js",
-    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install"
+    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install"
   },
   "exports": {
     ".": "./dist/index.js",

extensions/conversational-extension/src/index.ts

@@ -12,7 +12,7 @@ import {
  * functionality for managing threads.
  */
 export default class JSONConversationalExtension extends ConversationalExtension {
-  private static readonly _homeDir = 'file://threads'
+  private static readonly _threadFolder = 'file://threads'
   private static readonly _threadInfoFileName = 'thread.json'
   private static readonly _threadMessagesFileName = 'messages.jsonl'
 
@@ -20,8 +20,8 @@ export default class JSONConversationalExtension extends ConversationalExtension
    * Called when the extension is loaded.
    */
   async onLoad() {
-    if (!(await fs.existsSync(JSONConversationalExtension._homeDir)))
+    if (!(await fs.existsSync(JSONConversationalExtension._threadFolder)))
-      await fs.mkdirSync(JSONConversationalExtension._homeDir)
+      await fs.mkdirSync(JSONConversationalExtension._threadFolder)
     console.debug('JSONConversationalExtension loaded')
   }
 
@@ -68,7 +68,7 @@ export default class JSONConversationalExtension extends ConversationalExtension
   async saveThread(thread: Thread): Promise<void> {
     try {
       const threadDirPath = await joinPath([
-        JSONConversationalExtension._homeDir,
+        JSONConversationalExtension._threadFolder,
         thread.id,
       ])
       const threadJsonPath = await joinPath([
@@ -92,7 +92,7 @@ export default class JSONConversationalExtension extends ConversationalExtension
    */
   async deleteThread(threadId: string): Promise<void> {
     const path = await joinPath([
-      JSONConversationalExtension._homeDir,
+      JSONConversationalExtension._threadFolder,
       `${threadId}`,
     ])
     try {
@@ -109,7 +109,7 @@ export default class JSONConversationalExtension extends ConversationalExtension
   async addNewMessage(message: ThreadMessage): Promise<void> {
     try {
       const threadDirPath = await joinPath([
-        JSONConversationalExtension._homeDir,
+        JSONConversationalExtension._threadFolder,
         message.thread_id,
       ])
       const threadMessagePath = await joinPath([
@@ -177,7 +177,7 @@ export default class JSONConversationalExtension extends ConversationalExtension
   ): Promise<void> {
     try {
       const threadDirPath = await joinPath([
-        JSONConversationalExtension._homeDir,
+        JSONConversationalExtension._threadFolder,
         threadId,
       ])
       const threadMessagePath = await joinPath([
@@ -205,7 +205,7 @@ export default class JSONConversationalExtension extends ConversationalExtension
   private async readThread(threadDirName: string): Promise<any> {
     return fs.readFileSync(
       await joinPath([
-        JSONConversationalExtension._homeDir,
+        JSONConversationalExtension._threadFolder,
         threadDirName,
         JSONConversationalExtension._threadInfoFileName,
       ]),
@@ -219,14 +219,14 @@ export default class JSONConversationalExtension extends ConversationalExtension
    */
   private async getValidThreadDirs(): Promise<string[]> {
     const fileInsideThread: string[] = await fs.readdirSync(
-      JSONConversationalExtension._homeDir
+      JSONConversationalExtension._threadFolder
     )
 
     const threadDirs: string[] = []
     for (let i = 0; i < fileInsideThread.length; i++) {
       if (fileInsideThread[i].includes('.DS_Store')) continue
       const path = await joinPath([
-        JSONConversationalExtension._homeDir,
+        JSONConversationalExtension._threadFolder,
         fileInsideThread[i],
       ])
 
@@ -246,7 +246,7 @@ export default class JSONConversationalExtension extends ConversationalExtension
   async getAllMessages(threadId: string): Promise<ThreadMessage[]> {
     try {
       const threadDirPath = await joinPath([
-        JSONConversationalExtension._homeDir,
+        JSONConversationalExtension._threadFolder,
         threadId,
       ])
 
@@ -263,22 +263,17 @@ export default class JSONConversationalExtension extends ConversationalExtension
         JSONConversationalExtension._threadMessagesFileName,
       ])
 
-      const result = await fs
+      let readResult = await fs.readFileSync(messageFilePath, 'utf-8')
-        .readFileSync(messageFilePath, 'utf-8')
+
-        .then((content) =>
+      if (typeof readResult === 'object') {
-          content
+        readResult = JSON.stringify(readResult)
-            .toString()
+      }
-            .split('\n')
+
-            .filter((line) => line !== '')
+      const result = readResult.split('\n').filter((line) => line !== '')
-        )
 
       const messages: ThreadMessage[] = []
       result.forEach((line: string) => {
-        try {
+        messages.push(JSON.parse(line))
-          messages.push(JSON.parse(line) as ThreadMessage)
-        } catch (err) {
-          console.error(err)
-        }
       })
       return messages
     } catch (err) {

extensions/inference-nitro-extension/package.json

@@ -12,9 +12,9 @@
     "downloadnitro:darwin": "NITRO_VERSION=$(cat ./bin/version.txt) && download https://github.com/janhq/nitro/releases/download/v${NITRO_VERSION}/nitro-${NITRO_VERSION}-mac-arm64.tar.gz -e --strip 1 -o ./bin/mac-arm64 && chmod +x ./bin/mac-arm64/nitro && download https://github.com/janhq/nitro/releases/download/v${NITRO_VERSION}/nitro-${NITRO_VERSION}-mac-amd64.tar.gz -e --strip 1 -o ./bin/mac-x64 && chmod +x ./bin/mac-x64/nitro",
     "downloadnitro:win32": "download.bat",
     "downloadnitro": "run-script-os",
-    "build:publish:darwin": "rimraf *.tgz --glob && npm run build && npm run downloadnitro && ../../.github/scripts/auto-sign.sh && cpx \"bin/**\" \"dist/bin\" && npm pack && cpx *.tgz ../../electron/pre-install",
+    "build:publish:darwin": "rimraf *.tgz --glob && npm run build && npm run downloadnitro && ../../.github/scripts/auto-sign.sh && cpx \"bin/**\" \"dist/bin\" && npm pack && cpx *.tgz ../../pre-install",
-    "build:publish:win32": "rimraf *.tgz --glob && npm run build && npm run downloadnitro && cpx \"bin/**\" \"dist/bin\" && npm pack && cpx *.tgz ../../electron/pre-install",
+    "build:publish:win32": "rimraf *.tgz --glob && npm run build && npm run downloadnitro && cpx \"bin/**\" \"dist/bin\" && npm pack && cpx *.tgz ../../pre-install",
-    "build:publish:linux": "rimraf *.tgz --glob && npm run build && npm run downloadnitro && cpx \"bin/**\" \"dist/bin\" &&  npm pack && cpx *.tgz ../../electron/pre-install",
+    "build:publish:linux": "rimraf *.tgz --glob && npm run build && npm run downloadnitro && cpx \"bin/**\" \"dist/bin\" &&  npm pack && cpx *.tgz ../../pre-install",
     "build:publish": "run-script-os"
   },
   "exports": {

extensions/inference-nitro-extension/src/index.ts

@@ -243,9 +243,6 @@ export default class JanInferenceNitroExtension extends InferenceExtension {
    */
   private async onMessageRequest(data: MessageRequest) {
     if (data.model?.engine !== InferenceEngine.nitro || !this._currentModel) {
-      console.log(
-        `Model is not nitro or no model loaded ${data.model?.engine} ${this._currentModel}`
-      );
       return;
     }
 

extensions/inference-nitro-extension/src/node/execute.ts

@@ -25,12 +25,12 @@ export const executableNitroFile = (): NitroExecutableOptions => {
     if (nvidiaInfo["run_mode"] === "cpu") {
       binaryFolder = path.join(binaryFolder, "win-cpu");
     } else {
-      if (nvidiaInfo["cuda"].version === "12") {
+      if (nvidiaInfo["cuda"].version === "11") {
-        binaryFolder = path.join(binaryFolder, "win-cuda-12-0");
-      } else {
         binaryFolder = path.join(binaryFolder, "win-cuda-11-7");
+      } else {
+        binaryFolder = path.join(binaryFolder, "win-cuda-12-0");
       }
-      cudaVisibleDevices = nvidiaInfo["gpu_highest_vram"];
+      cudaVisibleDevices = nvidiaInfo["gpus_in_use"].join(",");
     }
     binaryName = "nitro.exe";
   } else if (process.platform === "darwin") {
@@ -50,12 +50,12 @@ export const executableNitroFile = (): NitroExecutableOptions => {
     if (nvidiaInfo["run_mode"] === "cpu") {
       binaryFolder = path.join(binaryFolder, "linux-cpu");
     } else {
-      if (nvidiaInfo["cuda"].version === "12") {
+      if (nvidiaInfo["cuda"].version === "11") {
-        binaryFolder = path.join(binaryFolder, "linux-cuda-12-0");
-      } else {
         binaryFolder = path.join(binaryFolder, "linux-cuda-11-7");
+      } else {
+        binaryFolder = path.join(binaryFolder, "linux-cuda-12-0");
       }
-      cudaVisibleDevices = nvidiaInfo["gpu_highest_vram"];
+      cudaVisibleDevices = nvidiaInfo["gpus_in_use"].join(",");
     }
   }
   return {

extensions/inference-nitro-extension/src/node/nvidia.ts

@@ -19,6 +19,8 @@ const DEFALT_SETTINGS = {
   },
   gpus: [],
   gpu_highest_vram: "",
+  gpus_in_use: [],
+  is_initial: true,
 };
 
 /**
@@ -48,11 +50,15 @@ export interface NitroProcessInfo {
  */
 export async function updateNvidiaInfo() {
   if (process.platform !== "darwin") {
-    await Promise.all([
+    let data;
-      updateNvidiaDriverInfo(),
+    try {
-      updateCudaExistence(),
+      data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
-      updateGpuInfo(),
+    } catch (error) {
-    ]);
+      data = DEFALT_SETTINGS;
+      writeFileSync(NVIDIA_INFO_FILE, JSON.stringify(data, null, 2));
+    }
+    updateNvidiaDriverInfo();
+    updateGpuInfo();
   }
 }
 
@@ -73,12 +79,7 @@ export async function updateNvidiaDriverInfo(): Promise<void> {
   exec(
     "nvidia-smi --query-gpu=driver_version --format=csv,noheader",
     (error, stdout) => {
-      let data;
+      let data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
-      try {
-        data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
-      } catch (error) {
-        data = DEFALT_SETTINGS;
-      }
 
       if (!error) {
         const firstLine = stdout.split("\n")[0].trim();
@@ -107,7 +108,7 @@ export function checkFileExistenceInPaths(
 /**
  * Validate cuda for linux and windows
  */
-export function updateCudaExistence() {
+export function updateCudaExistence(data: Record<string, any> = DEFALT_SETTINGS): Record<string, any> {
   let filesCuda12: string[];
   let filesCuda11: string[];
   let paths: string[];
@@ -141,19 +142,14 @@ export function updateCudaExistence() {
     cudaVersion = "12";
   }
 
-  let data;
-  try {
-    data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
-  } catch (error) {
-    data = DEFALT_SETTINGS;
-  }
-
   data["cuda"].exist = cudaExists;
   data["cuda"].version = cudaVersion;
-  if (cudaExists) {
+  console.log(data["is_initial"], data["gpus_in_use"]);
+  if (cudaExists && data["is_initial"] && data["gpus_in_use"].length > 0) {
     data.run_mode = "gpu";
   }
-  writeFileSync(NVIDIA_INFO_FILE, JSON.stringify(data, null, 2));
+  data.is_initial = false;
+  return data;
 }
 
 /**
@@ -161,14 +157,9 @@ export function updateCudaExistence() {
  */
 export async function updateGpuInfo(): Promise<void> {
   exec(
-    "nvidia-smi --query-gpu=index,memory.total --format=csv,noheader,nounits",
+    "nvidia-smi --query-gpu=index,memory.total,name --format=csv,noheader,nounits",
     (error, stdout) => {
-      let data;
+      let data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
-      try {
-        data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
-      } catch (error) {
-        data = DEFALT_SETTINGS;
-      }
 
       if (!error) {
         // Get GPU info and gpu has higher memory first
@@ -178,21 +169,27 @@ export async function updateGpuInfo(): Promise<void> {
           .trim()
           .split("\n")
           .map((line) => {
-            let [id, vram] = line.split(", ");
+            let [id, vram, name] = line.split(", ");
             vram = vram.replace(/\r/g, "");
             if (parseFloat(vram) > highestVram) {
               highestVram = parseFloat(vram);
               highestVramId = id;
             }
-            return { id, vram };
+            return { id, vram, name };
           });
 
-        data["gpus"] = gpus;
+        data.gpus = gpus;
-        data["gpu_highest_vram"] = highestVramId;
+        data.gpu_highest_vram = highestVramId;
       } else {
-        data["gpus"] = [];
+        data.gpus = [];
+        data.gpu_highest_vram = "";
       }
 
+      if (!data["gpus_in_use"] || data["gpus_in_use"].length === 0) {
+        data.gpus_in_use = [data["gpu_highest_vram"]];
+      }
+
+      data = updateCudaExistence(data);
       writeFileSync(NVIDIA_INFO_FILE, JSON.stringify(data, null, 2));
       Promise.resolve();
     }

extensions/inference-openai-extension/package.json

@@ -8,7 +8,7 @@
   "license": "AGPL-3.0",
   "scripts": {
     "build": "tsc -b . && webpack --config webpack.config.js",
-    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install"
+    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install"
   },
   "exports": {
     ".": "./dist/index.js",

extensions/inference-triton-trtllm-extension/package.json

@@ -8,7 +8,7 @@
   "license": "AGPL-3.0",
   "scripts": {
     "build": "tsc -b . && webpack --config webpack.config.js",
-    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install"
+    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install"
   },
   "exports": {
     ".": "./dist/index.js",

extensions/model-extension/package.json

@@ -8,7 +8,7 @@
   "license": "AGPL-3.0",
   "scripts": {
     "build": "tsc -b . && webpack --config webpack.config.js",
-    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install"
+    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install"
   },
   "devDependencies": {
     "cpx": "^1.5.0",

extensions/model-extension/src/@types/global.d.ts

@@ -1,3 +1,15 @@
-declare const EXTENSION_NAME: string
+export {}
-declare const MODULE_PATH: string
+declare global {
-declare const VERSION: stringå
+  declare const EXTENSION_NAME: string
+  declare const MODULE_PATH: string
+  declare const VERSION: string
+
+  interface Core {
+    api: APIFunctions
+    events: EventEmitter
+  }
+  interface Window {
+    core?: Core | undefined
+    electronAPI?: any | undefined
+  }
+}

extensions/model-extension/src/helpers/path.ts

@@ -0,0 +1,11 @@
+/**
+ *  try to retrieve the download file name from the source url
+ */
+
+export function extractFileName(url: string, fileExtension: string): string {
+  const extractedFileName = url.split('/').pop()
+  const fileName = extractedFileName.toLowerCase().endsWith(fileExtension)
+    ? extractedFileName
+    : extractedFileName + fileExtension
+  return fileName
+}

extensions/model-extension/src/index.ts

@@ -8,7 +8,13 @@ import {
   ModelExtension,
   Model,
   getJanDataFolderPath,
+  events,
+  DownloadEvent,
+  DownloadRoute,
+  ModelEvent,
 } from '@janhq/core'
+import { DownloadState } from '@janhq/core/.'
+import { extractFileName } from './helpers/path'
 
 /**
  * A extension for models
@@ -29,6 +35,8 @@ export default class JanModelExtension extends ModelExtension {
    */
   async onLoad() {
     this.copyModelsToHomeDir()
+    // Handle Desktop Events
+    this.handleDesktopEvents()
   }
 
   /**
@@ -61,6 +69,8 @@ export default class JanModelExtension extends ModelExtension {
 
       // Finished migration
       localStorage.setItem(`${EXTENSION_NAME}-version`, VERSION)
+
+      events.emit(ModelEvent.OnModelsUpdate, {})
     } catch (err) {
       console.error(err)
     }
@@ -83,31 +93,66 @@ export default class JanModelExtension extends ModelExtension {
     if (model.sources.length > 1) {
       // path to model binaries
       for (const source of model.sources) {
-        let path = this.extractFileName(source.url)
+        let path = extractFileName(
+          source.url,
+          JanModelExtension._supportedModelFormat
+        )
         if (source.filename) {
           path = await joinPath([modelDirPath, source.filename])
         }
 
         downloadFile(source.url, path, network)
       }
+      // TODO: handle multiple binaries for web later
     } else {
-      const fileName = this.extractFileName(model.sources[0]?.url)
+      const fileName = extractFileName(
+        model.sources[0]?.url,
+        JanModelExtension._supportedModelFormat
+      )
       const path = await joinPath([modelDirPath, fileName])
       downloadFile(model.sources[0]?.url, path, network)
+
+      if (window && window.core?.api && window.core.api.baseApiUrl) {
+        this.startPollingDownloadProgress(model.id)
+      }
     }
   }
 
   /**
-   *  try to retrieve the download file name from the source url
+   * Specifically for Jan server.
    */
-  private extractFileName(url: string): string {
+  private async startPollingDownloadProgress(modelId: string): Promise<void> {
-    const extractedFileName = url.split('/').pop()
+    // wait for some seconds before polling
-    const fileName = extractedFileName
+    await new Promise((resolve) => setTimeout(resolve, 3000))
-      .toLowerCase()
+
-      .endsWith(JanModelExtension._supportedModelFormat)
+    return new Promise((resolve) => {
-      ? extractedFileName
+      const interval = setInterval(async () => {
-      : extractedFileName + JanModelExtension._supportedModelFormat
+        fetch(
-    return fileName
+          `${window.core.api.baseApiUrl}/v1/download/${DownloadRoute.getDownloadProgress}/${modelId}`,
+          {
+            method: 'GET',
+            headers: { contentType: 'application/json' },
+          }
+        ).then(async (res) => {
+          const state: DownloadState = await res.json()
+          if (state.downloadState === 'end') {
+            events.emit(DownloadEvent.onFileDownloadSuccess, state)
+            clearInterval(interval)
+            resolve()
+            return
+          }
+
+          if (state.downloadState === 'error') {
+            events.emit(DownloadEvent.onFileDownloadError, state)
+            clearInterval(interval)
+            resolve()
+            return
+          }
+
+          events.emit(DownloadEvent.onFileDownloadUpdate, state)
+        })
+      }, 1000)
+    })
   }
 
   /**
@@ -286,6 +331,7 @@ export default class JanModelExtension extends ModelExtension {
    * model.json file associated with it.
    *
    * This function will create a model.json file for the model.
+   * It works only with single binary file model.
    *
    * @param dirName the director which reside in ~/jan/models but does not have model.json file.
    */
@@ -302,15 +348,14 @@ export default class JanModelExtension extends ModelExtension {
     let binaryFileSize: number | undefined = undefined
 
     for (const file of files) {
-      if (file.endsWith(JanModelExtension._incompletedModelFileName)) continue
+      if (file.endsWith(JanModelExtension._supportedModelFormat)) {
-      if (file.endsWith('.json')) continue
+        const path = await joinPath([JanModelExtension._homeDir, dirName, file])
-
+        const fileStats = await fs.fileStat(path)
-      const path = await joinPath([JanModelExtension._homeDir, dirName, file])
+        if (fileStats.isDirectory) continue
-      const fileStats = await fs.fileStat(path)
+        binaryFileSize = fileStats.size
-      if (fileStats.isDirectory) continue
+        binaryFileName = file
-      binaryFileSize = fileStats.size
+        break
-      binaryFileName = file
+      }
-      break
     }
 
     if (!binaryFileName) {
@@ -318,7 +363,7 @@ export default class JanModelExtension extends ModelExtension {
       return
     }
 
-    const defaultModel = await this.getDefaultModel()
+    const defaultModel = (await this.getDefaultModel()) as Model
     if (!defaultModel) {
       console.error('Unable to find default model')
       return
@@ -326,8 +371,19 @@ export default class JanModelExtension extends ModelExtension {
 
     const model: Model = {
       ...defaultModel,
+      // Overwrite default N/A fields
       id: dirName,
       name: dirName,
+      sources: [
+        {
+          url: binaryFileName,
+          filename: binaryFileName,
+        },
+      ],
+      settings: {
+        ...defaultModel.settings,
+        llama_model_path: binaryFileName,
+      },
       created: Date.now(),
       description: `${dirName} - user self import model`,
       metadata: {
@@ -371,4 +427,28 @@ export default class JanModelExtension extends ModelExtension {
   async getConfiguredModels(): Promise<Model[]> {
     return this.getModelsMetadata()
   }
+
+  handleDesktopEvents() {
+    if (window && window.electronAPI) {
+      window.electronAPI.onFileDownloadUpdate(
+        async (_event: string, state: any | undefined) => {
+          if (!state) return
+          state.downloadState = 'update'
+          events.emit(DownloadEvent.onFileDownloadUpdate, state)
+        }
+      )
+      window.electronAPI.onFileDownloadError(
+        async (_event: string, state: any) => {
+          state.downloadState = 'error'
+          events.emit(DownloadEvent.onFileDownloadError, state)
+        }
+      )
+      window.electronAPI.onFileDownloadSuccess(
+        async (_event: string, state: any) => {
+          state.downloadState = 'end'
+          events.emit(DownloadEvent.onFileDownloadSuccess, state)
+        }
+      )
+    }
+  }
 }

extensions/model-extension/tsconfig.json

@@ -8,7 +8,7 @@
     "forceConsistentCasingInFileNames": true,
     "strict": false,
     "skipLibCheck": true,
-    "rootDir": "./src"
+    "rootDir": "./src",
   },
-  "include": ["./src"]
+  "include": ["./src"],
 }

extensions/model-extension/webpack.config.js

@@ -19,7 +19,7 @@ module.exports = {
     new webpack.DefinePlugin({
       EXTENSION_NAME: JSON.stringify(packageJson.name),
       MODULE_PATH: JSON.stringify(`${packageJson.name}/${packageJson.module}`),
-      VERSION: JSON.stringify(packageJson.version),
+      VERSION: JSON.stringify(packageJson.version)
     }),
   ],
   output: {

extensions/monitoring-extension/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@janhq/monitoring-extension",
-  "version": "1.0.9",
+  "version": "1.0.10",
   "description": "This extension provides system health and OS level data",
   "main": "dist/index.js",
   "module": "dist/module.js",
@@ -8,7 +8,7 @@
   "license": "AGPL-3.0",
   "scripts": {
     "build": "tsc -b . && webpack --config webpack.config.js",
-    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../electron/pre-install"
+    "build:publish": "rimraf *.tgz --glob && npm run build && npm pack && cpx *.tgz ../../pre-install"
   },
   "devDependencies": {
     "rimraf": "^3.0.2",
@@ -26,6 +26,7 @@
     "README.md"
   ],
   "bundleDependencies": [
-    "node-os-utils"
+    "node-os-utils",
+    "@janhq/core"
   ]
 }

extensions/monitoring-extension/src/module.ts

@@ -1,4 +1,14 @@
 const nodeOsUtils = require("node-os-utils");
+const getJanDataFolderPath = require("@janhq/core/node").getJanDataFolderPath;
+const path = require("path");
+const { readFileSync } = require("fs");
+const exec = require("child_process").exec;
+
+const NVIDIA_INFO_FILE = path.join(
+  getJanDataFolderPath(),
+  "settings",
+  "settings.json"
+);
 
 const getResourcesInfo = () =>
   new Promise((resolve) => {
@@ -16,14 +26,44 @@ const getResourcesInfo = () =>
   });
 
 const getCurrentLoad = () =>
-  new Promise((resolve) => {
+  new Promise((resolve, reject) => {
     nodeOsUtils.cpu.usage().then((cpuPercentage) => {
-      const response = {
+      let data = {
-        cpu: {
+        run_mode: "cpu",
-          usage: cpuPercentage,
+        gpus_in_use: [],
-        },
       };
-      resolve(response);
+      if (process.platform !== "darwin") {
+        data = JSON.parse(readFileSync(NVIDIA_INFO_FILE, "utf-8"));
+      }
+      if (data.run_mode === "gpu" && data.gpus_in_use.length > 0) {
+        const gpuIds = data["gpus_in_use"].join(",");
+        if (gpuIds !== "") {
+          exec(
+            `nvidia-smi --query-gpu=index,name,temperature.gpu,utilization.gpu,memory.total,memory.free,utilization.memory --format=csv,noheader,nounits --id=${gpuIds}`,
+            (error, stdout, stderr) => {
+              if (error) {
+                console.error(`exec error: ${error}`);
+                reject(error);
+                return;
+              }
+              const gpuInfo = stdout.trim().split("\n").map((line) => {
+                const [id, name, temperature, utilization, memoryTotal, memoryFree, memoryUtilization] = line.split(", ").map(item => item.replace(/\r/g, ""));
+                return { id, name, temperature, utilization, memoryTotal, memoryFree, memoryUtilization };
+              });
+              resolve({
+                cpu: { usage: cpuPercentage },
+                gpu: gpuInfo
+              });
+            }
+          );
+        } else {
+          // Handle the case where gpuIds is empty
+          resolve({ cpu: { usage: cpuPercentage }, gpu: [] });
+        }
+      } else {
+        // Handle the case where run_mode is not 'gpu' or no GPUs are in use
+        resolve({ cpu: { usage: cpuPercentage }, gpu: [] });
+      }
     });
   });
 

package.json

@@ -21,22 +21,23 @@
     "lint": "yarn workspace jan lint && yarn workspace jan-web lint",
     "test:unit": "yarn workspace @janhq/core test",
     "test": "yarn workspace jan test:e2e",
-    "copy:assets": "cpx \"models/**\" \"electron/models/\" && cpx \"docs/openapi/**\" \"electron/docs/openapi\"",
+    "copy:assets": "cpx \"models/**\" \"electron/models/\" && cpx \"pre-install/*.tgz\" \"electron/pre-install/\" && cpx \"docs/openapi/**\" \"electron/docs/openapi\"",
     "dev:electron": "yarn copy:assets && yarn workspace jan dev",
     "dev:web": "yarn workspace jan-web dev",
-    "dev:server": "yarn workspace @janhq/server dev",
+    "dev:server": "yarn copy:assets && yarn workspace @janhq/server dev",
     "dev": "concurrently --kill-others \"yarn dev:web\" \"wait-on http://localhost:3000 && yarn dev:electron\"",
     "test-local": "yarn lint && yarn build:test && yarn test",
     "dev:uikit": "yarn workspace @janhq/uikit install && yarn workspace @janhq/uikit dev",
     "build:uikit": "yarn workspace @janhq/uikit install && yarn workspace @janhq/uikit build",
-    "build:server": "cd server && yarn install && yarn run build",
+    "build:server": "yarn copy:assets && cd server && yarn install && yarn run build",
     "build:core": "cd core && yarn install && yarn run build",
     "build:web": "yarn workspace jan-web build && cpx \"web/out/**\" \"electron/renderer/\"",
     "build:electron": "yarn copy:assets && yarn workspace jan build",
     "build:electron:test": "yarn workspace jan build:test",
-    "build:extensions:windows": "rimraf ./electron/pre-install/*.tgz && powershell -command \"$jobs = Get-ChildItem -Path './extensions' -Directory | ForEach-Object { Start-Job -Name ($_.Name) -ScriptBlock { param($_dir); try { Set-Location $_dir; npm install; npm run build:publish; Write-Output 'Build successful in ' + $_dir } catch { Write-Error 'Error in ' + $_dir; throw } } -ArgumentList $_.FullName }; $jobs | Wait-Job; $jobs | ForEach-Object { Receive-Job -Job $_ -Keep } | ForEach-Object { Write-Host $_ }; $failed = $jobs | Where-Object { $_.State -ne 'Completed' -or $_.ChildJobs[0].JobStateInfo.State -ne 'Completed' }; if ($failed) { Exit 1 }\"",
+    "build:extensions:windows": "rimraf ./pre-install/*.tgz && powershell -command \"$jobs = Get-ChildItem -Path './extensions' -Directory | ForEach-Object { Start-Job -Name ($_.Name) -ScriptBlock { param($_dir); try { Set-Location $_dir; npm install; npm run build:publish; Write-Output 'Build successful in ' + $_dir } catch { Write-Error 'Error in ' + $_dir; throw } } -ArgumentList $_.FullName }; $jobs | Wait-Job; $jobs | ForEach-Object { Receive-Job -Job $_ -Keep } | ForEach-Object { Write-Host $_ }; $failed = $jobs | Where-Object { $_.State -ne 'Completed' -or $_.ChildJobs[0].JobStateInfo.State -ne 'Completed' }; if ($failed) { Exit 1 }\"",
-    "build:extensions:linux": "rimraf ./electron/pre-install/*.tgz && find ./extensions -mindepth 1 -maxdepth 1 -type d -print0 | xargs -0 -n 1 -P 4 -I {} sh -c 'cd {} && npm install && npm run build:publish'",
+    "build:extensions:linux": "rimraf ./pre-install/*.tgz && find ./extensions -mindepth 1 -maxdepth 1 -type d -print0 | xargs -0 -n 1 -P 4 -I {} sh -c 'cd {} && npm install && npm run build:publish'",
-    "build:extensions:darwin": "rimraf ./electron/pre-install/*.tgz && find ./extensions -mindepth 1 -maxdepth 1 -type d -print0 | xargs -0 -n 1 -P 4 -I {} sh -c 'cd {} && npm install && npm run build:publish'",
+    "build:extensions:darwin": "rimraf ./pre-install/*.tgz && find ./extensions -mindepth 1 -maxdepth 1 -type d -print0 | xargs -0 -n 1 -P 4 -I {} sh -c 'cd {} && npm install && npm run build:publish'",
+    "build:extensions:server": "yarn workspace build:extensions ",
     "build:extensions": "run-script-os",
     "build:test": "yarn copy:assets && yarn build:web && yarn workspace jan build:test",
     "build": "yarn build:web && yarn build:electron",

server/helpers/setup.ts

@@ -0,0 +1,47 @@
+import { join, extname } from "path";
+import { existsSync, readdirSync, writeFileSync, mkdirSync } from "fs";
+import { init, installExtensions } from "@janhq/core/node";
+
+export async function setup() {
+  /**
+   * Setup Jan Data Directory
+   */
+  const appDir = process.env.JAN_DATA_DIRECTORY ?? join(__dirname, "..", "jan");
+
+  console.debug(`Create app data directory at ${appDir}...`);
+  if (!existsSync(appDir)) mkdirSync(appDir);
+  //@ts-ignore
+  global.core = {
+    // Define appPath function for app to retrieve app path globaly
+    appPath: () => appDir,
+  };
+  init({
+    extensionsPath: join(appDir, "extensions"),
+  });
+
+  /**
+   * Write app configurations. See #1619
+   */
+  console.debug("Writing config file...");
+  writeFileSync(
+    join(appDir, "settings.json"),
+    JSON.stringify({
+      data_folder: appDir,
+    }),
+    "utf-8"
+  );
+
+  /**
+   * Install extensions
+   */
+
+  console.debug("Installing extensions...");
+
+  const baseExtensionPath = join(__dirname, "../../..", "pre-install");
+  const extensions = readdirSync(baseExtensionPath)
+    .filter((file) => extname(file) === ".tgz")
+    .map((file) => join(baseExtensionPath, file));
+
+  await installExtensions(extensions);
+  console.debug("Extensions installed");
+}

server/index.ts

@@ -38,6 +38,7 @@ export interface ServerConfig {
   isVerboseEnabled?: boolean;
   schemaPath?: string;
   baseDir?: string;
+  storageAdataper?: any;
 }
 
 /**
@@ -103,9 +104,12 @@ export const startServer = async (configs?: ServerConfig) => {
       { prefix: "extensions" }
     );
 
+    // Register proxy middleware
+    if (configs?.storageAdataper)
+      server.addHook("preHandler", configs.storageAdataper);
+
     // Register API routes
     await server.register(v1Router, { prefix: "/v1" });
-
     // Start listening for requests
     await server
       .listen({

server/main.ts

@@ -1,3 +1,7 @@
-import { startServer } from "./index";
+import { s3 } from "./middleware/s3";
-
+import { setup } from "./helpers/setup";
-startServer();
+import { startServer as start } from "./index";
+/**
+ * Setup extensions and start the server
+ */
+setup().then(() => start({ storageAdataper: s3 }));

server/middleware/s3.ts

@@ -0,0 +1,70 @@
+import { join } from "path";
+
+// Middleware to intercept requests and proxy if certain conditions are met
+const config = {
+  endpoint: process.env.AWS_ENDPOINT,
+  region: process.env.AWS_REGION,
+  credentials: {
+    accessKeyId: process.env.AWS_ACCESS_KEY_ID,
+    secretAccessKey: process.env.AWS_SECRET_ACCESS_KEY,
+  },
+};
+
+const S3_BUCKET_NAME = process.env.S3_BUCKET_NAME;
+
+const fs = require("@cyclic.sh/s3fs")(S3_BUCKET_NAME, config);
+const PROXY_PREFIX = "/v1/fs";
+const PROXY_ROUTES = ["/threads", "/messages"];
+
+export const s3 = (req: any, reply: any, done: any) => {
+  // Proxy FS requests to S3 using S3FS
+  if (req.url.startsWith(PROXY_PREFIX)) {
+    const route = req.url.split("/").pop();
+    const args = parseRequestArgs(req);
+
+    // Proxy matched requests to the s3fs module
+    if (args.length && PROXY_ROUTES.some((route) => args[0].includes(route))) {
+      try {
+        // Handle customized route
+        // S3FS does not handle appendFileSync
+        if (route === "appendFileSync") {
+          let result = handAppendFileSync(args);
+
+          reply.status(200).send(result);
+          return;
+        }
+        // Reroute the other requests to the s3fs module
+        const result = fs[route](...args);
+        reply.status(200).send(result);
+        return;
+      } catch (ex) {
+        console.log(ex);
+      }
+    }
+  }
+  // Let other requests go through
+  done();
+};
+
+const parseRequestArgs = (req: Request) => {
+  const {
+    getJanDataFolderPath,
+    normalizeFilePath,
+  } = require("@janhq/core/node");
+
+  return JSON.parse(req.body as any).map((arg: any) =>
+    typeof arg === "string" &&
+    (arg.startsWith(`file:/`) || arg.startsWith(`file:\\`))
+      ? join(getJanDataFolderPath(), normalizeFilePath(arg))
+      : arg
+  );
+};
+
+const handAppendFileSync = (args: any[]) => {
+  if (fs.existsSync(args[0])) {
+    const data = fs.readFileSync(args[0], "utf-8");
+    return fs.writeFileSync(args[0], data + args[1]);
+  } else {
+    return fs.writeFileSync(args[0], args[1]);
+  }
+};

server/nodemon.json

@@ -1,5 +0,0 @@
-{
-    "watch": ["main.ts", "v1"],
-    "ext": "ts, json",
-    "exec": "tsc && node ./build/main.js"
-}

server/package.json

@@ -13,16 +13,18 @@
   "scripts": {
     "lint": "eslint . --ext \".js,.jsx,.ts,.tsx\"",
     "test:e2e": "playwright test --workers=1",
-    "dev": "tsc --watch & node --watch build/main.js",
+    "build:core": "cd node_modules/@janhq/core && yarn install && yarn build",
-    "build": "tsc"
+    "dev": "yarn build:core && tsc --watch & node --watch build/main.js",
+    "build": "yarn build:core && tsc"
   },
   "dependencies": {
     "@alumna/reflect": "^1.1.3",
+    "@cyclic.sh/s3fs": "^1.2.9",
     "@fastify/cors": "^8.4.2",
     "@fastify/static": "^6.12.0",
     "@fastify/swagger": "^8.13.0",
     "@fastify/swagger-ui": "2.0.1",
-    "@janhq/core": "link:./core",
+    "@janhq/core": "file:../core",
     "dotenv": "^16.3.1",
     "fastify": "^4.24.3",
     "request": "^2.88.2",
@@ -39,5 +41,8 @@
     "run-script-os": "^1.1.6",
     "@types/tcp-port-used": "^1.0.4",
     "typescript": "^5.2.2"
-  }
+  },
+  "bundleDependencies": [
+    "@janhq/core"
+  ]
 }

server/tsconfig.json

@@ -20,5 +20,5 @@
   // "sourceMap": true,
 
   "include": ["./**/*.ts"],
-  "exclude": ["core", "build", "dist", "tests", "node_modules"]
+  "exclude": ["core", "build", "dist", "tests", "node_modules", "extensions"]
 }

web/containers/Layout/BottomBar/DownloadingState/index.tsx

@@ -13,22 +13,22 @@ import {
 import { useAtomValue } from 'jotai'
 
 import useDownloadModel from '@/hooks/useDownloadModel'
-import { useDownloadState } from '@/hooks/useDownloadState'
+import { modelDownloadStateAtom } from '@/hooks/useDownloadState'
 
 import { formatDownloadPercentage } from '@/utils/converter'
 
-import { downloadingModelsAtom } from '@/helpers/atoms/Model.atom'
+import { getDownloadingModelAtom } from '@/helpers/atoms/Model.atom'
 
 export default function DownloadingState() {
-  const { downloadStates } = useDownloadState()
+  const downloadStates = useAtomValue(modelDownloadStateAtom)
-  const downloadingModels = useAtomValue(downloadingModelsAtom)
+  const downloadingModels = useAtomValue(getDownloadingModelAtom)
   const { abortModelDownload } = useDownloadModel()
 
-  const totalCurrentProgress = downloadStates
+  const totalCurrentProgress = Object.values(downloadStates)
     .map((a) => a.size.transferred + a.size.transferred)
     .reduce((partialSum, a) => partialSum + a, 0)
 
-  const totalSize = downloadStates
+  const totalSize = Object.values(downloadStates)
     .map((a) => a.size.total + a.size.total)
     .reduce((partialSum, a) => partialSum + a, 0)
 
@@ -36,12 +36,14 @@ export default function DownloadingState() {
 
   return (
     <Fragment>
-      {downloadStates?.length > 0 && (
+      {Object.values(downloadStates)?.length > 0 && (
         <Modal>
           <ModalTrigger asChild>
             <div className="relative block">
               <Button size="sm" themes="outline">
-                <span>{downloadStates.length} Downloading model</span>
+                <span>
+                  {Object.values(downloadStates).length} Downloading model
+                </span>
               </Button>
               <span
                 className="absolute left-0 h-full rounded-md rounded-l-md bg-primary/20"
@@ -55,40 +57,38 @@ export default function DownloadingState() {
             <ModalHeader>
               <ModalTitle>Downloading model</ModalTitle>
             </ModalHeader>
-            {downloadStates.map((item, i) => {
+            {Object.values(downloadStates).map((item, i) => (
-              return (
+              <div className="pt-2" key={i}>
-                <div className="pt-2" key={i}>
+                <Progress
-                  <Progress
+                  className="mb-2 h-2"
-                    className="mb-2 h-2"
+                  value={
-                    value={
+                    formatDownloadPercentage(item?.percent, {
-                      formatDownloadPercentage(item?.percent, {
+                      hidePercentage: true,
-                        hidePercentage: true,
+                    }) as number
-                      }) as number
+                  }
-                    }
+                />
-                  />
+                <div className="flex items-center justify-between gap-x-2">
-                  <div className="flex items-center justify-between gap-x-2">
+                  <div className="flex gap-x-2">
-                    <div className="flex gap-x-2">
+                    <p className="line-clamp-1">{item?.modelId}</p>
-                      <p className="line-clamp-1">{item?.modelId}</p>
+                    <span>{formatDownloadPercentage(item?.percent)}</span>
-                      <span>{formatDownloadPercentage(item?.percent)}</span>
-                    </div>
-                    <Button
-                      themes="outline"
-                      size="sm"
-                      onClick={() => {
-                        if (item?.modelId) {
-                          const model = downloadingModels.find(
-                            (model) => model.id === item.modelId
-                          )
-                          if (model) abortModelDownload(model)
-                        }
-                      }}
-                    >
-                      Cancel
-                    </Button>
                   </div>
+                  <Button
+                    themes="outline"
+                    size="sm"
+                    onClick={() => {
+                      if (item?.modelId) {
+                        const model = downloadingModels.find(
+                          (model) => model.id === item.modelId
+                        )
+                        if (model) abortModelDownload(model)
+                      }
+                    }}
+                  >
+                    Cancel
+                  </Button>
                 </div>
-              )
+              </div>
-            })}
+            ))}
           </ModalContent>
         </Modal>
       )}

web/containers/Layout/BottomBar/index.tsx

@@ -25,12 +25,12 @@ import { MainViewState } from '@/constants/screens'
 
 import { useActiveModel } from '@/hooks/useActiveModel'
 
-import { useDownloadState } from '@/hooks/useDownloadState'
+import { modelDownloadStateAtom } from '@/hooks/useDownloadState'
-import { useGetDownloadedModels } from '@/hooks/useGetDownloadedModels'
 import useGetSystemResources from '@/hooks/useGetSystemResources'
 import { useMainViewState } from '@/hooks/useMainViewState'
 
 import { serverEnabledAtom } from '@/helpers/atoms/LocalServer.atom'
+import { downloadedModelsAtom } from '@/helpers/atoms/Model.atom'
 
 const menuLinks = [
   {
@@ -47,14 +47,22 @@ const menuLinks = [
 
 const BottomBar = () => {
   const { activeModel, stateModel } = useActiveModel()
-  const { ram, cpu } = useGetSystemResources()
+  const { ram, cpu, gpus } = useGetSystemResources()
   const progress = useAtomValue(appDownloadProgress)
-  const { downloadedModels } = useGetDownloadedModels()
+  const downloadedModels = useAtomValue(downloadedModelsAtom)
+
   const { setMainViewState } = useMainViewState()
-  const { downloadStates } = useDownloadState()
+  const downloadStates = useAtomValue(modelDownloadStateAtom)
   const setShowSelectModelModal = useSetAtom(showSelectModelModalAtom)
   const [serverEnabled] = useAtom(serverEnabledAtom)
 
+  const calculateGpuMemoryUsage = (gpu: Record<string, never>) => {
+    const total = parseInt(gpu.memoryTotal)
+    const free = parseInt(gpu.memoryFree)
+    if (!total || !free) return 0
+    return Math.round(((total - free) / total) * 100)
+  }
+
   return (
     <div className="fixed bottom-0 left-16 z-20 flex h-12 w-[calc(100%-64px)] items-center justify-between border-t border-border bg-background/80 px-3">
       <div className="flex flex-shrink-0 items-center gap-x-2">
@@ -100,7 +108,7 @@ const BottomBar = () => {
           )}
         {downloadedModels.length === 0 &&
           !stateModel.loading &&
-          downloadStates.length === 0 && (
+          Object.values(downloadStates).length === 0 && (
             <Button
               size="sm"
               themes="outline"
@@ -117,6 +125,17 @@ const BottomBar = () => {
           <SystemItem name="CPU:" value={`${cpu}%`} />
           <SystemItem name="Mem:" value={`${ram}%`} />
         </div>
+        {gpus.length > 0 && (
+          <div className="flex items-center gap-x-2">
+            {gpus.map((gpu, index) => (
+              <SystemItem
+                key={index}
+                name={`GPU ${gpu.id}:`}
+                value={`${gpu.utilization}% Util, ${calculateGpuMemoryUsage(gpu)}% Mem`}
+              />
+            ))}
+          </div>
+        )}
         {/* VERSION is defined by webpack, please see next.config.js */}
         <span className="text-xs text-muted-foreground">
           Jan v{VERSION ?? ''}

web/containers/Layout/TopBar/CommandListDownloadedModel/index.tsx

@@ -11,7 +11,7 @@ import {
   Badge,
 } from '@janhq/uikit'
 
-import { useAtom } from 'jotai'
+import { useAtom, useAtomValue } from 'jotai'
 import { DatabaseIcon, CpuIcon } from 'lucide-react'
 
 import { showSelectModelModalAtom } from '@/containers/Providers/KeyListener'
@@ -19,14 +19,14 @@ import { showSelectModelModalAtom } from '@/containers/Providers/KeyListener'
 import { MainViewState } from '@/constants/screens'
 
 import { useActiveModel } from '@/hooks/useActiveModel'
-import { useGetDownloadedModels } from '@/hooks/useGetDownloadedModels'
 import { useMainViewState } from '@/hooks/useMainViewState'
 
 import { serverEnabledAtom } from '@/helpers/atoms/LocalServer.atom'
+import { downloadedModelsAtom } from '@/helpers/atoms/Model.atom'
 
 export default function CommandListDownloadedModel() {
   const { setMainViewState } = useMainViewState()
-  const { downloadedModels } = useGetDownloadedModels()
+  const downloadedModels = useAtomValue(downloadedModelsAtom)
   const { activeModel, startModel, stopModel } = useActiveModel()
   const [serverEnabled] = useAtom(serverEnabledAtom)
   const [showSelectModelModal, setShowSelectModelModal] = useAtom(

web/containers/Layout/TopBar/index.tsx

@@ -20,7 +20,6 @@ import { MainViewState } from '@/constants/screens'
 
 import { useClickOutside } from '@/hooks/useClickOutside'
 import { useCreateNewThread } from '@/hooks/useCreateNewThread'
-import useGetAssistants, { getAssistants } from '@/hooks/useGetAssistants'
 import { useMainViewState } from '@/hooks/useMainViewState'
 
 import { usePath } from '@/hooks/usePath'
@@ -29,13 +28,14 @@ import { showRightSideBarAtom } from '@/screens/Chat/Sidebar'
 
 import { openFileTitle } from '@/utils/titleUtils'
 
+import { assistantsAtom } from '@/helpers/atoms/Assistant.atom'
 import { activeThreadAtom } from '@/helpers/atoms/Thread.atom'
 
 const TopBar = () => {
   const activeThread = useAtomValue(activeThreadAtom)
   const { mainViewState } = useMainViewState()
   const { requestCreateNewThread } = useCreateNewThread()
-  const { assistants } = useGetAssistants()
+  const assistants = useAtomValue(assistantsAtom)
   const [showRightSideBar, setShowRightSideBar] = useAtom(showRightSideBarAtom)
   const [showLeftSideBar, setShowLeftSideBar] = useAtom(showLeftSideBarAtom)
   const showing = useAtomValue(showRightSideBarAtom)
@@ -61,12 +61,7 @@ const TopBar = () => {
 
   const onCreateConversationClick = async () => {
     if (assistants.length === 0) {
-      const res = await getAssistants()
+      alert('No assistant available')
-      if (res.length === 0) {
-        alert('No assistant available')
-        return
-      }
-      requestCreateNewThread(res[0])
     } else {
       requestCreateNewThread(assistants[0])
     }

web/containers/ModalCancelDownload/index.tsx

@@ -17,23 +17,22 @@ import {
 import { atom, useAtomValue } from 'jotai'
 
 import useDownloadModel from '@/hooks/useDownloadModel'
-import { useDownloadState } from '@/hooks/useDownloadState'
+
+import { modelDownloadStateAtom } from '@/hooks/useDownloadState'
 
 import { formatDownloadPercentage } from '@/utils/converter'
 
-import { downloadingModelsAtom } from '@/helpers/atoms/Model.atom'
+import { getDownloadingModelAtom } from '@/helpers/atoms/Model.atom'
 
 type Props = {
   model: Model
   isFromList?: boolean
 }
 
-export default function ModalCancelDownload({ model, isFromList }: Props) {
+const ModalCancelDownload: React.FC<Props> = ({ model, isFromList }) => {
-  const { modelDownloadStateAtom } = useDownloadState()
+  const downloadingModels = useAtomValue(getDownloadingModelAtom)
-  const downloadingModels = useAtomValue(downloadingModelsAtom)
   const downloadAtom = useMemo(
     () => atom((get) => get(modelDownloadStateAtom)[model.id]),
-    // eslint-disable-next-line react-hooks/exhaustive-deps
     [model.id]
   )
   const downloadState = useAtomValue(downloadAtom)
@@ -98,3 +97,5 @@ export default function ModalCancelDownload({ model, isFromList }: Props) {
     </Modal>
   )
 }
+
+export default ModalCancelDownload

web/containers/Providers/DataLoader.tsx

@@ -0,0 +1,21 @@
+'use client'
+
+import { Fragment, ReactNode } from 'react'
+
+import useAssistants from '@/hooks/useAssistants'
+import useModels from '@/hooks/useModels'
+import useThreads from '@/hooks/useThreads'
+
+type Props = {
+  children: ReactNode
+}
+
+const DataLoader: React.FC<Props> = ({ children }) => {
+  useModels()
+  useThreads()
+  useAssistants()
+
+  return <Fragment>{children}</Fragment>
+}
+
+export default DataLoader

web/containers/Providers/EventHandler.tsx

@@ -18,7 +18,6 @@ import {
   loadModelErrorAtom,
   stateModelAtom,
 } from '@/hooks/useActiveModel'
-import { useGetDownloadedModels } from '@/hooks/useGetDownloadedModels'
 
 import { queuedMessageAtom } from '@/hooks/useSendChatMessage'
 
@@ -29,16 +28,18 @@ import {
   addNewMessageAtom,
   updateMessageAtom,
 } from '@/helpers/atoms/ChatMessage.atom'
+import { downloadedModelsAtom } from '@/helpers/atoms/Model.atom'
 import {
   updateThreadWaitingForResponseAtom,
   threadsAtom,
   isGeneratingResponseAtom,
+  updateThreadAtom,
 } from '@/helpers/atoms/Thread.atom'
 
 export default function EventHandler({ children }: { children: ReactNode }) {
   const addNewMessage = useSetAtom(addNewMessageAtom)
   const updateMessage = useSetAtom(updateMessageAtom)
-  const { downloadedModels } = useGetDownloadedModels()
+  const downloadedModels = useAtomValue(downloadedModelsAtom)
   const setActiveModel = useSetAtom(activeModelAtom)
   const setStateModel = useSetAtom(stateModelAtom)
   const setQueuedMessage = useSetAtom(queuedMessageAtom)
@@ -49,6 +50,7 @@ export default function EventHandler({ children }: { children: ReactNode }) {
   const modelsRef = useRef(downloadedModels)
   const threadsRef = useRef(threads)
   const setIsGeneratingResponse = useSetAtom(isGeneratingResponseAtom)
+  const updateThread = useSetAtom(updateThreadAtom)
 
   useEffect(() => {
     threadsRef.current = threads
@@ -126,11 +128,17 @@ export default function EventHandler({ children }: { children: ReactNode }) {
 
       const thread = threadsRef.current?.find((e) => e.id == message.thread_id)
       if (thread) {
-        const messageContent = message.content[0]?.text.value ?? ''
+        const messageContent = message.content[0]?.text?.value
         const metadata = {
           ...thread.metadata,
-          lastMessage: messageContent,
+          ...(messageContent && { lastMessage: messageContent }),
         }
+
+        updateThread({
+          ...thread,
+          metadata,
+        })
+
         extensionManager
           .get<ConversationalExtension>(ExtensionTypeEnum.Conversational)
           ?.saveThread({
@@ -143,7 +151,7 @@ export default function EventHandler({ children }: { children: ReactNode }) {
           ?.addNewMessage(message)
       }
     },
-    [updateMessage, updateThreadWaiting]
+    [updateMessage, updateThreadWaiting, setIsGeneratingResponse, updateThread]
   )
 
   useEffect(() => {

web/containers/Providers/EventListener.tsx

@@ -1,91 +1,62 @@
 /* eslint-disable @typescript-eslint/no-explicit-any */
+import { PropsWithChildren, useCallback, useEffect } from 'react'
 
-import { PropsWithChildren, useEffect, useRef } from 'react'
+import React from 'react'
 
-import { baseName } from '@janhq/core'
+import { DownloadEvent, events } from '@janhq/core'
-import { useAtomValue, useSetAtom } from 'jotai'
+import { useSetAtom } from 'jotai'
 
-import { useDownloadState } from '@/hooks/useDownloadState'
+import { setDownloadStateAtom } from '@/hooks/useDownloadState'
-import { useGetDownloadedModels } from '@/hooks/useGetDownloadedModels'
-
-import { modelBinFileName } from '@/utils/model'
 
 import EventHandler from './EventHandler'
 
 import { appDownloadProgress } from './Jotai'
 
-import { downloadingModelsAtom } from '@/helpers/atoms/Model.atom'
+const EventListenerWrapper = ({ children }: PropsWithChildren) => {
-
+  const setDownloadState = useSetAtom(setDownloadStateAtom)
-export default function EventListenerWrapper({ children }: PropsWithChildren) {
   const setProgress = useSetAtom(appDownloadProgress)
-  const models = useAtomValue(downloadingModelsAtom)
-  const modelsRef = useRef(models)
 
-  const { setDownloadedModels, downloadedModels } = useGetDownloadedModels()
+  const onFileDownloadUpdate = useCallback(
-  const {
+    async (state: DownloadState) => {
-    setDownloadState,
+      console.debug('onFileDownloadUpdate', state)
-    setDownloadStateSuccess,
+      setDownloadState(state)
-    setDownloadStateFailed,
+    },
-    setDownloadStateCancelled,
+    [setDownloadState]
-  } = useDownloadState()
+  )
-  const downloadedModelRef = useRef(downloadedModels)
+
+  const onFileDownloadError = useCallback(
+    (state: DownloadState) => {
+      console.debug('onFileDownloadError', state)
+      setDownloadState(state)
+    },
+    [setDownloadState]
+  )
+
+  const onFileDownloadSuccess = useCallback(
+    (state: DownloadState) => {
+      console.debug('onFileDownloadSuccess', state)
+      setDownloadState(state)
+    },
+    [setDownloadState]
+  )
 
   useEffect(() => {
-    modelsRef.current = models
+    console.log('EventListenerWrapper: registering event listeners...')
-  }, [models])
+
-  useEffect(() => {
+    events.on(DownloadEvent.onFileDownloadUpdate, onFileDownloadUpdate)
-    downloadedModelRef.current = downloadedModels
+    events.on(DownloadEvent.onFileDownloadError, onFileDownloadError)
-  }, [downloadedModels])
+    events.on(DownloadEvent.onFileDownloadSuccess, onFileDownloadSuccess)
+
+    return () => {
+      console.log('EventListenerWrapper: unregistering event listeners...')
+      events.off(DownloadEvent.onFileDownloadUpdate, onFileDownloadUpdate)
+      events.off(DownloadEvent.onFileDownloadError, onFileDownloadError)
+      events.off(DownloadEvent.onFileDownloadSuccess, onFileDownloadSuccess)
+    }
+  }, [onFileDownloadUpdate, onFileDownloadError, onFileDownloadSuccess])
 
   useEffect(() => {
     if (window && window.electronAPI) {
-      window.electronAPI.onFileDownloadUpdate(
-        async (_event: string, state: any | undefined) => {
-          if (!state) return
-          const modelName = await baseName(state.fileName)
-          const model = modelsRef.current.find(
-            (model) => modelBinFileName(model) === modelName
-          )
-          if (model)
-            setDownloadState({
-              ...state,
-              modelId: model.id,
-            })
-        }
-      )
-
-      window.electronAPI.onFileDownloadError(
-        async (_event: string, state: any) => {
-          const modelName = await baseName(state.fileName)
-          const model = modelsRef.current.find(
-            (model) => modelBinFileName(model) === modelName
-          )
-          if (model) {
-            if (state.err?.message !== 'aborted') {
-              console.error('Download error', state)
-              setDownloadStateFailed(model.id, state.err.message)
-            } else {
-              setDownloadStateCancelled(model.id)
-            }
-          }
-        }
-      )
-
-      window.electronAPI.onFileDownloadSuccess(
-        async (_event: string, state: any) => {
-          if (state && state.fileName) {
-            const modelName = await baseName(state.fileName)
-            const model = modelsRef.current.find(
-              (model) => modelBinFileName(model) === modelName
-            )
-            if (model) {
-              setDownloadStateSuccess(model.id)
-              setDownloadedModels([...downloadedModelRef.current, model])
-            }
-          }
-        }
-      )
-
       window.electronAPI.onAppUpdateDownloadUpdate(
         (_event: string, progress: any) => {
           setProgress(progress.percent)
@@ -105,14 +76,9 @@ export default function EventListenerWrapper({ children }: PropsWithChildren) {
       })
     }
     return () => {}
-  }, [
+  }, [setDownloadState, setProgress])
-    setDownloadState,
-    setDownloadStateCancelled,
-    setDownloadStateFailed,
-    setDownloadStateSuccess,
-    setDownloadedModels,
-    setProgress,
-  ])
 
   return <EventHandler>{children}</EventHandler>
 }
+
+export default EventListenerWrapper

web/containers/Providers/index.tsx

@@ -21,6 +21,10 @@ import {
 
 import Umami from '@/utils/umami'
 
+import Loader from '../Loader'
+
+import DataLoader from './DataLoader'
+
 import KeyListener from './KeyListener'
 
 import { extensionManager } from '@/extension'
@@ -30,6 +34,7 @@ const Providers = (props: PropsWithChildren) => {
 
   const [setupCore, setSetupCore] = useState(false)
   const [activated, setActivated] = useState(false)
+  const [settingUp, setSettingUp] = useState(false)
 
   async function setupExtensions() {
     // Register all active extensions
@@ -37,11 +42,13 @@ const Providers = (props: PropsWithChildren) => {
 
     setTimeout(async () => {
       if (!isCoreExtensionInstalled()) {
-        setupBaseExtensions()
+        setSettingUp(true)
+        await setupBaseExtensions()
         return
       }
 
       extensionManager.load()
+      setSettingUp(false)
       setActivated(true)
     }, 500)
   }
@@ -71,11 +78,14 @@ const Providers = (props: PropsWithChildren) => {
     <JotaiWrapper>
       <ThemeWrapper>
         <Umami />
+        {settingUp && <Loader description="Preparing Update..." />}
         {setupCore && activated && (
           <KeyListener>
             <FeatureToggleWrapper>
               <EventListenerWrapper>
-                <TooltipProvider delayDuration={0}>{children}</TooltipProvider>
+                <TooltipProvider delayDuration={0}>
+                  <DataLoader>{children}</DataLoader>
+                </TooltipProvider>
                 {!isMac && <GPUDriverPrompt />}
               </EventListenerWrapper>
               <Toaster />

web/helpers/atoms/Assistant.atom.ts

@@ -0,0 +1,4 @@
+import { Assistant } from '@janhq/core/.'
+import { atom } from 'jotai'
+
+export const assistantsAtom = atom<Assistant[]>([])

web/helpers/atoms/ChatMessage.atom.ts

@@ -70,11 +70,12 @@ export const addNewMessageAtom = atom(
     set(chatMessages, newData)
 
     // Update thread last message
-    set(
+    if (newMessage.content.length)
-      updateThreadStateLastMessageAtom,
+      set(
-      newMessage.thread_id,
+        updateThreadStateLastMessageAtom,
-      newMessage.content
+        newMessage.thread_id,
-    )
+        newMessage.content
+      )
   }
 )
 
@@ -131,7 +132,8 @@ export const updateMessageAtom = atom(
       newData[conversationId] = updatedMessages
       set(chatMessages, newData)
       // Update thread last message
-      set(updateThreadStateLastMessageAtom, conversationId, text)
+      if (text.length)
+        set(updateThreadStateLastMessageAtom, conversationId, text)
     }
   }
 )

web/helpers/atoms/Model.atom.ts

@@ -4,23 +4,32 @@ import { atom } from 'jotai'
 export const stateModel = atom({ state: 'start', loading: false, model: '' })
 export const activeAssistantModelAtom = atom<Model | undefined>(undefined)
 
-export const downloadingModelsAtom = atom<Model[]>([])
+/**
+ * Stores the list of models which are being downloaded.
+ */
+const downloadingModelsAtom = atom<Model[]>([])
 
-export const addNewDownloadingModelAtom = atom(
+export const getDownloadingModelAtom = atom((get) => get(downloadingModelsAtom))
-  null,
+
-  (get, set, model: Model) => {
+export const addDownloadingModelAtom = atom(null, (get, set, model: Model) => {
-    const currentModels = get(downloadingModelsAtom)
+  const downloadingModels = get(downloadingModelsAtom)
-    set(downloadingModelsAtom, [...currentModels, model])
+  if (!downloadingModels.find((e) => e.id === model.id)) {
+    set(downloadingModelsAtom, [...downloadingModels, model])
   }
-)
+})
 
 export const removeDownloadingModelAtom = atom(
   null,
   (get, set, modelId: string) => {
-    const currentModels = get(downloadingModelsAtom)
+    const downloadingModels = get(downloadingModelsAtom)
+
     set(
       downloadingModelsAtom,
-      currentModels.filter((e) => e.id !== modelId)
+      downloadingModels.filter((e) => e.id !== modelId)
     )
   }
 )
+
+export const downloadedModelsAtom = atom<Model[]>([])
+
+export const configuredModelsAtom = atom<Model[]>([])

web/helpers/atoms/SystemBar.atom.ts

@@ -5,3 +5,5 @@ export const usedRamAtom = atom<number>(0)
 export const availableRamAtom = atom<number>(0)
 
 export const cpuUsageAtom = atom<number>(0)
+
+export const nvidiaTotalVramAtom = atom<number>(0)

web/hooks/useActiveModel.ts

@@ -3,9 +3,9 @@ import { atom, useAtom, useAtomValue, useSetAtom } from 'jotai'
 
 import { toaster } from '@/containers/Toast'
 
-import { useGetDownloadedModels } from './useGetDownloadedModels'
 import { LAST_USED_MODEL_ID } from './useRecommendedModel'
 
+import { downloadedModelsAtom } from '@/helpers/atoms/Model.atom'
 import { activeThreadAtom } from '@/helpers/atoms/Thread.atom'
 
 export const activeModelAtom = atom<Model | undefined>(undefined)
@@ -21,7 +21,7 @@ export function useActiveModel() {
   const [activeModel, setActiveModel] = useAtom(activeModelAtom)
   const activeThread = useAtomValue(activeThreadAtom)
   const [stateModel, setStateModel] = useAtom(stateModelAtom)
-  const { downloadedModels } = useGetDownloadedModels()
+  const downloadedModels = useAtomValue(downloadedModelsAtom)
   const setLoadModelError = useSetAtom(loadModelErrorAtom)
 
   const startModel = async (modelId: string) => {

web/hooks/useAssistants.ts

@@ -0,0 +1,39 @@
+import { useCallback, useEffect } from 'react'
+
+import {
+  Assistant,
+  AssistantEvent,
+  AssistantExtension,
+  ExtensionTypeEnum,
+  events,
+} from '@janhq/core'
+
+import { useSetAtom } from 'jotai'
+
+import { extensionManager } from '@/extension'
+import { assistantsAtom } from '@/helpers/atoms/Assistant.atom'
+
+const useAssistants = () => {
+  const setAssistants = useSetAtom(assistantsAtom)
+
+  const getData = useCallback(async () => {
+    const assistants = await getLocalAssistants()
+    setAssistants(assistants)
+  }, [setAssistants])
+
+  useEffect(() => {
+    getData()
+
+    events.on(AssistantEvent.OnAssistantsUpdate, () => getData())
+    return () => {
+      events.off(AssistantEvent.OnAssistantsUpdate, () => getData())
+    }
+  }, [getData])
+}
+
+const getLocalAssistants = async (): Promise<Assistant[]> =>
+  extensionManager
+    .get<AssistantExtension>(ExtensionTypeEnum.Assistant)
+    ?.getAssistants() ?? []
+
+export default useAssistants

web/hooks/useCreateNewThread.ts

@@ -6,8 +6,9 @@ import {
   ThreadAssistantInfo,
   ThreadState,
   Model,
+  MessageStatus,
 } from '@janhq/core'
-import { atom, useSetAtom } from 'jotai'
+import { atom, useAtomValue, useSetAtom } from 'jotai'
 
 import { selectedModelAtom } from '@/containers/DropdownListSidebar'
 import { fileUploadAtom } from '@/containers/Providers/Jotai'
@@ -19,6 +20,8 @@ import useRecommendedModel from './useRecommendedModel'
 import useSetActiveThread from './useSetActiveThread'
 
 import { extensionManager } from '@/extension'
+
+import { getCurrentChatMessagesAtom } from '@/helpers/atoms/ChatMessage.atom'
 import {
   threadsAtom,
   threadStatesAtom,
@@ -50,15 +53,25 @@ export const useCreateNewThread = () => {
   const setFileUpload = useSetAtom(fileUploadAtom)
   const setSelectedModel = useSetAtom(selectedModelAtom)
   const setThreadModelParams = useSetAtom(setThreadModelParamsAtom)
+  const messages = useAtomValue(getCurrentChatMessagesAtom)
 
   const { recommendedModel, downloadedModels } = useRecommendedModel()
 
+  const threads = useAtomValue(threadsAtom)
+
   const requestCreateNewThread = async (
     assistant: Assistant,
     model?: Model | undefined
   ) => {
     const defaultModel = model ?? recommendedModel ?? downloadedModels[0]
 
+    // check last thread message, if there empty last message use can not create thread
+    const lastMessage = threads[threads.length - 1]?.metadata?.lastMessage
+
+    if (!lastMessage && threads.length && !messages.length) {
+      return null
+    }
+
     const createdAt = Date.now()
     const assistantInfo: ThreadAssistantInfo = {
       assistant_id: assistant.id,

web/hooks/useDeleteModel.ts

@@ -1,13 +1,14 @@
 import { ExtensionTypeEnum, ModelExtension, Model } from '@janhq/core'
 
+import { useAtom } from 'jotai'
+
 import { toaster } from '@/containers/Toast'
 
-import { useGetDownloadedModels } from '@/hooks/useGetDownloadedModels'
-
 import { extensionManager } from '@/extension/ExtensionManager'
+import { downloadedModelsAtom } from '@/helpers/atoms/Model.atom'
 
 export default function useDeleteModel() {
-  const { setDownloadedModels, downloadedModels } = useGetDownloadedModels()
+  const [downloadedModels, setDownloadedModels] = useAtom(downloadedModelsAtom)
 
   const deleteModel = async (model: Model) => {
     await extensionManager

web/hooks/useDownloadModel.ts

@@ -1,4 +1,4 @@
-import { useContext } from 'react'
+import { useCallback, useContext } from 'react'
 
 import {
   Model,
@@ -15,21 +15,40 @@ import { FeatureToggleContext } from '@/context/FeatureToggle'
 
 import { modelBinFileName } from '@/utils/model'
 
-import { useDownloadState } from './useDownloadState'
+import { setDownloadStateAtom } from './useDownloadState'
 
 import { extensionManager } from '@/extension/ExtensionManager'
-import { addNewDownloadingModelAtom } from '@/helpers/atoms/Model.atom'
+import { addDownloadingModelAtom } from '@/helpers/atoms/Model.atom'
 
 export default function useDownloadModel() {
   const { ignoreSSL, proxy } = useContext(FeatureToggleContext)
-  const { setDownloadState } = useDownloadState()
+  const setDownloadState = useSetAtom(setDownloadStateAtom)
-  const addNewDownloadingModel = useSetAtom(addNewDownloadingModelAtom)
+  const addDownloadingModel = useSetAtom(addDownloadingModelAtom)
 
-  const downloadModel = async (model: Model) => {
+  const downloadModel = useCallback(
-    const childrenDownloadProgress: DownloadState[] = []
+    async (model: Model) => {
-    model.sources.forEach((source: ModelArtifact) => {
+      const childProgresses: DownloadState[] = model.sources.map(
-      childrenDownloadProgress.push({
+        (source: ModelArtifact) => ({
-        modelId: source.filename,
+          filename: source.filename,
+          modelId: model.id,
+          time: {
+            elapsed: 0,
+            remaining: 0,
+          },
+          speed: 0,
+          percent: 0,
+          size: {
+            total: 0,
+            transferred: 0,
+          },
+          downloadState: 'downloading',
+        })
+      )
+
+      // set an initial download state
+      setDownloadState({
+        filename: '',
+        modelId: model.id,
         time: {
           elapsed: 0,
           remaining: 0,
@@ -40,31 +59,16 @@ export default function useDownloadModel() {
           total: 0,
           transferred: 0,
         },
+        children: childProgresses,
+        downloadState: 'downloading',
       })
-    })
 
-    // set an initial download state
+      addDownloadingModel(model)
-    setDownloadState({
-      modelId: model.id,
-      time: {
-        elapsed: 0,
-        remaining: 0,
-      },
-      speed: 0,
-      percent: 0,
-      size: {
-        total: 0,
-        transferred: 0,
-      },
-      children: childrenDownloadProgress,
-    })
 
-    addNewDownloadingModel(model)
+      await localDownloadModel(model, ignoreSSL, proxy)
-
+    },
-    await extensionManager
+    [ignoreSSL, proxy, addDownloadingModel, setDownloadState]
-      .get<ModelExtension>(ExtensionTypeEnum.Model)
+  )
-      ?.downloadModel(model, { ignoreSSL, proxy })
-  }
 
   const abortModelDownload = async (model: Model) => {
     await abortDownload(
@@ -77,3 +81,12 @@ export default function useDownloadModel() {
     abortModelDownload,
   }
 }
+
+const localDownloadModel = async (
+  model: Model,
+  ignoreSSL: boolean,
+  proxy: string
+) =>
+  extensionManager
+    .get<ModelExtension>(ExtensionTypeEnum.Model)
+    ?.downloadModel(model, { ignoreSSL, proxy })

web/hooks/useDownloadState.ts

@@ -1,96 +1,64 @@
-import { atom, useSetAtom, useAtomValue } from 'jotai'
+import { atom } from 'jotai'
 
 import { toaster } from '@/containers/Toast'
 
+import {
+  configuredModelsAtom,
+  downloadedModelsAtom,
+  removeDownloadingModelAtom,
+} from '@/helpers/atoms/Model.atom'
+
 // download states
-const modelDownloadStateAtom = atom<Record<string, DownloadState>>({})
+export const modelDownloadStateAtom = atom<Record<string, DownloadState>>({})
 
-const setDownloadStateAtom = atom(null, (get, set, state: DownloadState) => {
+/**
-  const currentState = { ...get(modelDownloadStateAtom) }
+ * Used to set the download state for a particular model.
-  console.debug(
+ */
-    `current download state for ${state.modelId} is ${JSON.stringify(state)}`
+export const setDownloadStateAtom = atom(
-  )
-  currentState[state.modelId] = state
-  set(modelDownloadStateAtom, currentState)
-})
-
-const setDownloadStateSuccessAtom = atom(null, (get, set, modelId: string) => {
-  const currentState = { ...get(modelDownloadStateAtom) }
-  const state = currentState[modelId]
-  if (!state) {
-    console.debug(`Cannot find download state for ${modelId}`)
-    return
-  }
-  delete currentState[modelId]
-  set(modelDownloadStateAtom, currentState)
-  toaster({
-    title: 'Download Completed',
-    description: `Download ${modelId} completed`,
-    type: 'success',
-  })
-})
-
-const setDownloadStateFailedAtom = atom(
   null,
-  (get, set, modelId: string, error: string) => {
+  (get, set, state: DownloadState) => {
     const currentState = { ...get(modelDownloadStateAtom) }
-    const state = currentState[modelId]
-    if (!state) {
-      console.debug(`Cannot find download state for ${modelId}`)
-      return
-    }
-    if (error.includes('certificate')) {
-      error += '. To fix enable "Ignore SSL Certificates" in Advanced settings.'
-    }
-    toaster({
-      title: 'Download Failed',
-      description: `Model ${modelId} download failed: ${error}`,
-      type: 'error',
-    })
 
-    delete currentState[modelId]
+    if (state.downloadState === 'end') {
-    set(modelDownloadStateAtom, currentState)
+      // download successfully
-  }
+      delete currentState[state.modelId]
-)
+      set(removeDownloadingModelAtom, state.modelId)
-const setDownloadStateCancelledAtom = atom(
+      const model = get(configuredModelsAtom).find(
-  null,
+        (e) => e.id === state.modelId
-  (get, set, modelId: string) => {
+      )
-    const currentState = { ...get(modelDownloadStateAtom) }
+      if (model) set(downloadedModelsAtom, (prev) => [...prev, model])
-    const state = currentState[modelId]
-    if (!state) {
-      console.debug(`Cannot find download state for ${modelId}`)
       toaster({
-        title: 'Cancel Download',
+        title: 'Download Completed',
-        description: `Model ${modelId} cancel download`,
+        description: `Download ${state.modelId} completed`,
-        type: 'warning',
+        type: 'success',
       })
-
+    } else if (state.downloadState === 'error') {
-      return
+      // download error
+      delete currentState[state.modelId]
+      set(removeDownloadingModelAtom, state.modelId)
+      if (state.error === 'aborted') {
+        toaster({
+          title: 'Cancel Download',
+          description: `Model ${state.modelId} download cancelled`,
+          type: 'warning',
+        })
+      } else {
+        let error = state.error
+        if (state.error?.includes('certificate')) {
+          error +=
+            '. To fix enable "Ignore SSL Certificates" in Advanced settings.'
+        }
+        toaster({
+          title: 'Download Failed',
+          description: `Model ${state.modelId} download failed: ${error}`,
+          type: 'error',
+        })
+      }
+    } else {
+      // download in progress
+      currentState[state.modelId] = state
     }
-    delete currentState[modelId]
+
     set(modelDownloadStateAtom, currentState)
   }
 )
-
-export function useDownloadState() {
-  const modelDownloadState = useAtomValue(modelDownloadStateAtom)
-  const setDownloadState = useSetAtom(setDownloadStateAtom)
-  const setDownloadStateSuccess = useSetAtom(setDownloadStateSuccessAtom)
-  const setDownloadStateFailed = useSetAtom(setDownloadStateFailedAtom)
-  const setDownloadStateCancelled = useSetAtom(setDownloadStateCancelledAtom)
-
-  const downloadStates: DownloadState[] = []
-  for (const [, value] of Object.entries(modelDownloadState)) {
-    downloadStates.push(value)
-  }
-
-  return {
-    modelDownloadStateAtom,
-    modelDownloadState,
-    setDownloadState,
-    setDownloadStateSuccess,
-    setDownloadStateFailed,
-    setDownloadStateCancelled,
-    downloadStates,
-  }
-}

web/hooks/useGetAssistants.ts

@@ -1,27 +0,0 @@
-import { useEffect, useState } from 'react'
-
-import { Assistant, ExtensionTypeEnum, AssistantExtension } from '@janhq/core'
-
-import { extensionManager } from '@/extension/ExtensionManager'
-
-export const getAssistants = async (): Promise<Assistant[]> =>
-  extensionManager
-    .get<AssistantExtension>(ExtensionTypeEnum.Assistant)
-    ?.getAssistants() ?? []
-
-/**
- * Hooks for get assistants
- *
- * @returns assistants
- */
-export default function useGetAssistants() {
-  const [assistants, setAssistants] = useState<Assistant[]>([])
-
-  useEffect(() => {
-    getAssistants()
-      .then((data) => setAssistants(data))
-      .catch((err) => console.error(err))
-  }, [])
-
-  return { assistants }
-}

web/hooks/useGetConfiguredModels.ts

@@ -1,30 +0,0 @@
-import { useCallback, useEffect, useState } from 'react'
-
-import { ExtensionTypeEnum, ModelExtension, Model } from '@janhq/core'
-
-import { extensionManager } from '@/extension/ExtensionManager'
-
-export function useGetConfiguredModels() {
-  const [loading, setLoading] = useState<boolean>(false)
-  const [models, setModels] = useState<Model[]>([])
-
-  const fetchModels = useCallback(async () => {
-    setLoading(true)
-    const models = await getConfiguredModels()
-    setLoading(false)
-    setModels(models)
-  }, [])
-
-  useEffect(() => {
-    fetchModels()
-  }, [fetchModels])
-
-  return { loading, models }
-}
-
-const getConfiguredModels = async (): Promise<Model[]> => {
-  const models = await extensionManager
-    .get<ModelExtension>(ExtensionTypeEnum.Model)
-    ?.getConfiguredModels()
-  return models ?? []
-}