ci: update path-filtering orb to 1.3.0 (#3588)

Signed-off-by: Jared Van Bortel <jared@nomic.ai>

.circleci/config.yml

@@ -1,7 +1,7 @@
 version: 2.1
 setup: true
 orbs:
-  path-filtering: circleci/path-filtering@1.1.0
+  path-filtering: circleci/path-filtering@1.3.0
 
 workflows:
   version: 2.1

.circleci/continue_config.yml

@@ -18,6 +18,68 @@ parameters:
     type: boolean
     default: false
 
+job-macos-executor: &job-macos-executor
+  macos:
+    xcode: 16.2.0
+  resource_class: macos.m1.medium.gen1
+  environment:
+    HOMEBREW_NO_AUTO_UPDATE: 1
+
+job-macos-install-deps: &job-macos-install-deps
+  name: Install basic macOS build dependencies
+  command: brew install ccache llvm wget
+
+job-linux-install-chat-deps: &job-linux-install-chat-deps
+  name: Install Linux build dependencies for gpt4all-chat
+  command: |
+    # Prevent apt-get from interactively prompting for service restart
+    echo "\$nrconf{restart} = 'l'" | sudo tee /etc/needrestart/conf.d/90-autorestart.conf >/dev/null
+    wget -qO- 'https://apt.llvm.org/llvm-snapshot.gpg.key' | sudo tee /etc/apt/trusted.gpg.d/apt.llvm.org.asc >/dev/null
+    sudo add-apt-repository -yn 'deb http://apt.llvm.org/jammy/ llvm-toolchain-jammy-19 main'
+    wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" \
+      | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc >/dev/null
+    wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" \
+      | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list >/dev/null
+    wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
+    sudo dpkg -i cuda-keyring_1.1-1_all.deb
+    packages=(
+      bison build-essential ccache clang-19 clang-tools-19 cuda-compiler-11-8 flex gperf libcublas-dev-11-8
+      libfontconfig1 libfreetype6 libgl1-mesa-dev libmysqlclient21 libnvidia-compute-550-server libodbc2 libpq5
+      libstdc++-12-dev libwayland-dev libx11-6 libx11-xcb1 libxcb-cursor0 libxcb-glx0 libxcb-icccm4 libxcb-image0
+      libxcb-keysyms1 libxcb-randr0 libxcb-render-util0 libxcb-shape0 libxcb-shm0 libxcb-sync1 libxcb-util1
+      libxcb-xfixes0 libxcb-xinerama0 libxcb-xkb1 libxcb1 libxext6 libxfixes3 libxi6 libxkbcommon-dev libxkbcommon-x11-0
+      libxrender1 patchelf python3 vulkan-sdk python3 vulkan-sdk
+    )
+    sudo apt-get update
+    sudo apt-get install -y "${packages[@]}"
+    wget "https://qt.mirror.constant.com/archive/online_installers/4.8/qt-online-installer-linux-x64-4.8.1.run"
+    chmod +x qt-online-installer-linux-x64-4.8.1.run
+    ./qt-online-installer-linux-x64-4.8.1.run --no-force-installations --no-default-installations \
+      --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations \
+      --email "$QT_EMAIL" --password "$QT_PASSWORD" install \
+      qt.tools.cmake qt.tools.ifw.48 qt.tools.ninja qt.qt6.682.linux_gcc_64 qt.qt6.682.addons.qt5compat \
+      qt.qt6.682.debug_info extensions.qtpdf.682 qt.qt6.682.addons.qthttpserver
+
+job-linux-install-backend-deps: &job-linux-install-backend-deps
+  name: Install Linux build dependencies for gpt4all-backend
+  command: |
+    wget -qO- 'https://apt.llvm.org/llvm-snapshot.gpg.key' | sudo tee /etc/apt/trusted.gpg.d/apt.llvm.org.asc >/dev/null
+    sudo add-apt-repository -yn 'deb http://apt.llvm.org/jammy/ llvm-toolchain-jammy-19 main'
+    wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" \
+      | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc >/dev/null
+    wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" \
+      | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list >/dev/null
+    wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
+    sudo dpkg -i cuda-keyring_1.1-1_all.deb
+    packages=(
+      build-essential ccache clang-19 clang-tools-19 cuda-compiler-11-8 libcublas-dev-11-8
+      libnvidia-compute-550-server libstdc++-12-dev vulkan-sdk
+    )
+    sudo apt-get update
+    sudo apt-get install -y "${packages[@]}"
+    pyenv global 3.13.2
+    pip install setuptools wheel cmake ninja
+
 jobs:
   # work around CircleCI-Public/path-filtering-orb#20
   noop:
@@ -34,8 +96,7 @@ jobs:
           name: Verify that commit is on the main branch
           command: git merge-base --is-ancestor HEAD main
   build-offline-chat-installer-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - run:
@@ -46,12 +107,11 @@ jobs:
       - restore_cache:
           keys:
             - ccache-gpt4all-macos-
+      - run:
+          <<: *job-macos-install-deps
       - run:
           name: Install Rosetta
           command: softwareupdate --install-rosetta --agree-to-license  # needed for QtIFW
-      - run:
-          name: Install dependencies
-          command: brew install ccache wget
       - run:
           name: Installing Qt
           command: |
@@ -86,6 +146,9 @@ jobs:
               -DCMAKE_BUILD_TYPE=Release \
               -DCMAKE_PREFIX_PATH:PATH=~/Qt/6.8.2/macos/lib/cmake \
               -DCMAKE_MAKE_PROGRAM:FILEPATH=~/Qt/Tools/Ninja/ninja \
+              -DCMAKE_C_COMPILER=/opt/homebrew/opt/llvm/bin/clang \
+              -DCMAKE_CXX_COMPILER=/opt/homebrew/opt/llvm/bin/clang++ \
+              -DCMAKE_RANLIB=/usr/bin/ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DBUILD_UNIVERSAL=ON \
@@ -128,8 +191,7 @@ jobs:
             - upload
 
   sign-offline-chat-installer-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       # attach to a workspace containing unsigned dmg
@@ -163,8 +225,7 @@ jobs:
             - upload
 
   notarize-offline-chat-installer-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - attach_workspace:
@@ -203,8 +264,7 @@ jobs:
             hdiutil detach /Volumes/gpt4all-installer-darwin
 
   build-online-chat-installer-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - run:
@@ -215,12 +275,11 @@ jobs:
       - restore_cache:
           keys:
             - ccache-gpt4all-macos-
+      - run:
+          <<: *job-macos-install-deps
       - run:
           name: Install Rosetta
           command: softwareupdate --install-rosetta --agree-to-license  # needed for QtIFW
-      - run:
-          name: Install dependencies
-          command: brew install ccache wget
       - run:
           name: Installing Qt
           command: |
@@ -255,6 +314,9 @@ jobs:
               -DCMAKE_BUILD_TYPE=Release \
               -DCMAKE_PREFIX_PATH:PATH=~/Qt/6.8.2/macos/lib/cmake \
               -DCMAKE_MAKE_PROGRAM:FILEPATH=~/Qt/Tools/Ninja/ninja \
+              -DCMAKE_C_COMPILER=/opt/homebrew/opt/llvm/bin/clang \
+              -DCMAKE_CXX_COMPILER=/opt/homebrew/opt/llvm/bin/clang++ \
+              -DCMAKE_RANLIB=/usr/bin/ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DBUILD_UNIVERSAL=ON \
@@ -290,8 +352,7 @@ jobs:
             - upload
 
   sign-online-chat-installer-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       # attach to a workspace containing unsigned dmg
@@ -325,8 +386,7 @@ jobs:
             - upload
 
   notarize-online-chat-installer-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - attach_workspace:
@@ -379,34 +439,7 @@ jobs:
           keys:
             - ccache-gpt4all-linux-amd64-
       - run:
-          name: Setup Linux and Dependencies
+          <<: *job-linux-install-chat-deps
-          command: |
-            # Prevent apt-get from interactively prompting for service restart
-            echo "\$nrconf{restart} = 'l'" | sudo tee /etc/needrestart/conf.d/90-autorestart.conf >/dev/null
-            wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc
-            wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list
-            wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
-            sudo dpkg -i cuda-keyring_1.1-1_all.deb
-            packages=(
-              bison build-essential ccache cuda-compiler-11-8 flex g++-12 gperf libcublas-dev-11-8 libfontconfig1
-              libfreetype6 libgl1-mesa-dev libmysqlclient21 libnvidia-compute-550-server libodbc2 libpq5 libwayland-dev
-              libx11-6 libx11-xcb1 libxcb-cursor0 libxcb-glx0 libxcb-icccm4 libxcb-image0 libxcb-keysyms1 libxcb-randr0
-              libxcb-render-util0 libxcb-shape0 libxcb-shm0 libxcb-sync1 libxcb-util1 libxcb-xfixes0 libxcb-xinerama0
-              libxcb-xkb1 libxcb1 libxext6 libxfixes3 libxi6 libxkbcommon-dev libxkbcommon-x11-0 libxrender1 patchelf
-              python3 vulkan-sdk
-            )
-            sudo apt-get update
-            sudo apt-get install -y "${packages[@]}"
-      - run:
-          name: Installing Qt
-          command: |
-            wget "https://qt.mirror.constant.com/archive/online_installers/4.8/qt-online-installer-linux-x64-4.8.1.run"
-            chmod +x qt-online-installer-linux-x64-4.8.1.run
-            ./qt-online-installer-linux-x64-4.8.1.run --no-force-installations --no-default-installations \
-              --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations \
-              --email "$QT_EMAIL" --password "$QT_PASSWORD" install \
-              qt.tools.cmake qt.tools.ifw.48 qt.tools.ninja qt.qt6.682.linux_gcc_64 qt.qt6.682.addons.qt5compat \
-              qt.qt6.682.debug_info extensions.qtpdf.682 qt.qt6.682.addons.qthttpserver
       - run:
           name: Build linuxdeployqt
           command: |
@@ -427,8 +460,10 @@ jobs:
             ~/Qt/Tools/CMake/bin/cmake \
               -S ../gpt4all-chat -B . \
               -DCMAKE_BUILD_TYPE=Release \
-              -DCMAKE_C_COMPILER=gcc-12 \
+              -DCMAKE_C_COMPILER=clang-19 \
-              -DCMAKE_CXX_COMPILER=g++-12 \
+              -DCMAKE_CXX_COMPILER=clang++-19 \
+              -DCMAKE_CXX_COMPILER_AR=ar \
+              -DCMAKE_CXX_COMPILER_RANLIB=ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CUDA_COMPILER_LAUNCHER=ccache \
@@ -468,34 +503,7 @@ jobs:
           keys:
             - ccache-gpt4all-linux-amd64-
       - run:
-          name: Setup Linux and Dependencies
+          <<: *job-linux-install-chat-deps
-          command: |
-            # Prevent apt-get from interactively prompting for service restart
-            echo "\$nrconf{restart} = 'l'" | sudo tee /etc/needrestart/conf.d/90-autorestart.conf >/dev/null
-            wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc
-            wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list
-            wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
-            sudo dpkg -i cuda-keyring_1.1-1_all.deb
-            packages=(
-              bison build-essential ccache cuda-compiler-11-8 flex g++-12 gperf libcublas-dev-11-8 libfontconfig1
-              libfreetype6 libgl1-mesa-dev libmysqlclient21 libnvidia-compute-550-server libodbc2 libpq5 libwayland-dev
-              libx11-6 libx11-xcb1 libxcb-cursor0 libxcb-glx0 libxcb-icccm4 libxcb-image0 libxcb-keysyms1 libxcb-randr0
-              libxcb-render-util0 libxcb-shape0 libxcb-shm0 libxcb-sync1 libxcb-util1 libxcb-xfixes0 libxcb-xinerama0
-              libxcb-xkb1 libxcb1 libxext6 libxfixes3 libxi6 libxkbcommon-dev libxkbcommon-x11-0 libxrender1 patchelf
-              python3 vulkan-sdk
-            )
-            sudo apt-get update
-            sudo apt-get install -y "${packages[@]}"
-      - run:
-          name: Installing Qt
-          command: |
-            wget "https://qt.mirror.constant.com/archive/online_installers/4.8/qt-online-installer-linux-x64-4.8.1.run"
-            chmod +x qt-online-installer-linux-x64-4.8.1.run
-            ./qt-online-installer-linux-x64-4.8.1.run --no-force-installations --no-default-installations \
-              --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations \
-              --email "$QT_EMAIL" --password "$QT_PASSWORD" install \
-              qt.tools.cmake qt.tools.ifw.48 qt.tools.ninja qt.qt6.682.linux_gcc_64 qt.qt6.682.addons.qt5compat \
-              qt.qt6.682.debug_info extensions.qtpdf.682 qt.qt6.682.addons.qthttpserver
       - run:
           name: Build linuxdeployqt
           command: |
@@ -516,8 +524,10 @@ jobs:
             ~/Qt/Tools/CMake/bin/cmake \
               -S ../gpt4all-chat -B . \
               -DCMAKE_BUILD_TYPE=Release \
-              -DCMAKE_C_COMPILER=gcc-12 \
+              -DCMAKE_C_COMPILER=clang-19 \
-              -DCMAKE_CXX_COMPILER=g++-12 \
+              -DCMAKE_CXX_COMPILER=clang++-19 \
+              -DCMAKE_CXX_COMPILER_AR=ar \
+              -DCMAKE_CXX_COMPILER_RANLIB=ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CUDA_COMPILER_LAUNCHER=ccache \
@@ -1109,34 +1119,7 @@ jobs:
           keys:
             - ccache-gpt4all-linux-amd64-
       - run:
-          name: Setup Linux and Dependencies
+          <<: *job-linux-install-chat-deps
-          command: |
-            # Prevent apt-get from interactively prompting for service restart
-            echo "\$nrconf{restart} = 'l'" | sudo tee /etc/needrestart/conf.d/90-autorestart.conf >/dev/null
-            wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc
-            wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list
-            wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
-            sudo dpkg -i cuda-keyring_1.1-1_all.deb
-            packages=(
-              bison build-essential ccache cuda-compiler-11-8 flex g++-12 gperf libcublas-dev-11-8 libfontconfig1
-              libfreetype6 libgl1-mesa-dev libmysqlclient21 libnvidia-compute-550-server libodbc2 libpq5 libwayland-dev
-              libx11-6 libx11-xcb1 libxcb-cursor0 libxcb-glx0 libxcb-icccm4 libxcb-image0 libxcb-keysyms1 libxcb-randr0
-              libxcb-render-util0 libxcb-shape0 libxcb-shm0 libxcb-sync1 libxcb-util1 libxcb-xfixes0 libxcb-xinerama0
-              libxcb-xkb1 libxcb1 libxext6 libxfixes3 libxi6 libxkbcommon-dev libxkbcommon-x11-0 libxrender1 python3
-              vulkan-sdk
-            )
-            sudo apt-get update
-            sudo apt-get install -y "${packages[@]}"
-      - run:
-          name: Installing Qt
-          command: |
-            wget "https://qt.mirror.constant.com/archive/online_installers/4.8/qt-online-installer-linux-x64-4.8.1.run"
-            chmod +x qt-online-installer-linux-x64-4.8.1.run
-            ./qt-online-installer-linux-x64-4.8.1.run --no-force-installations --no-default-installations \
-              --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations \
-              --email "$QT_EMAIL" --password "$QT_PASSWORD" install \
-              qt.tools.cmake qt.tools.ifw.48 qt.tools.ninja qt.qt6.682.linux_gcc_64 qt.qt6.682.addons.qt5compat \
-              qt.qt6.682.debug_info extensions.qtpdf.682 qt.qt6.682.addons.qthttpserver
       - run:
           name: Build
           no_output_timeout: 30m
@@ -1147,8 +1130,10 @@ jobs:
             ~/Qt/Tools/CMake/bin/cmake \
               -S gpt4all-chat -B build \
               -DCMAKE_BUILD_TYPE=Release \
-              -DCMAKE_C_COMPILER=gcc-12 \
+              -DCMAKE_C_COMPILER=clang-19 \
-              -DCMAKE_CXX_COMPILER=g++-12 \
+              -DCMAKE_CXX_COMPILER=clang++-19 \
+              -DCMAKE_CXX_COMPILER_AR=ar \
+              -DCMAKE_CXX_COMPILER_RANLIB=ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CUDA_COMPILER_LAUNCHER=ccache \
@@ -1227,8 +1212,7 @@ jobs:
             - ..\.ccache
 
   build-gpt4all-chat-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - run:
@@ -1239,12 +1223,11 @@ jobs:
       - restore_cache:
           keys:
             - ccache-gpt4all-macos-
+      - run:
+          <<: *job-macos-install-deps
       - run:
           name: Install Rosetta
           command: softwareupdate --install-rosetta --agree-to-license  # needed for QtIFW
-      - run:
-          name: Install dependencies
-          command: brew install ccache wget
       - run:
           name: Installing Qt
           command: |
@@ -1267,6 +1250,9 @@ jobs:
               -DCMAKE_BUILD_TYPE=Release \
               -DCMAKE_PREFIX_PATH:PATH=~/Qt/6.8.2/macos/lib/cmake \
               -DCMAKE_MAKE_PROGRAM:FILEPATH=~/Qt/Tools/Ninja/ninja \
+              -DCMAKE_C_COMPILER=/opt/homebrew/opt/llvm/bin/clang \
+              -DCMAKE_CXX_COMPILER=/opt/homebrew/opt/llvm/bin/clang++ \
+              -DCMAKE_RANLIB=/usr/bin/ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DBUILD_UNIVERSAL=ON \
@@ -1334,22 +1320,7 @@ jobs:
           keys:
             - ccache-gpt4all-linux-amd64-
       - run:
-          name: Set Python Version
+          <<: *job-linux-install-backend-deps
-          command: pyenv global 3.11.2
-      - run:
-          name: Install dependencies
-          command: |
-            wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc
-            wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list
-            wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
-            sudo dpkg -i cuda-keyring_1.1-1_all.deb
-            packages=(
-              build-essential ccache cmake cuda-compiler-11-8 g++-12 libcublas-dev-11-8 libnvidia-compute-550-server
-              vulkan-sdk
-            )
-            sudo apt-get update
-            sudo apt-get install -y "${packages[@]}"
-            pip install setuptools wheel cmake
       - run:
           name: Build C library
           no_output_timeout: 30m
@@ -1358,15 +1329,17 @@ jobs:
             git submodule update --init --recursive
             ccache -o "cache_dir=${PWD}/../.ccache" -o max_size=500M -p -z
             cd gpt4all-backend
-            cmake -B build \
+            cmake -B build -G Ninja \
               -DCMAKE_BUILD_TYPE=Release \
-              -DCMAKE_C_COMPILER=gcc-12 \
+              -DCMAKE_C_COMPILER=clang-19 \
-              -DCMAKE_CXX_COMPILER=g++-12 \
+              -DCMAKE_CXX_COMPILER=clang++-19 \
+              -DCMAKE_CXX_COMPILER_AR=ar \
+              -DCMAKE_CXX_COMPILER_RANLIB=ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CUDA_COMPILER_LAUNCHER=ccache \
               -DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON \
-              -DCMAKE_CUDA_ARCHITECTURES='52-virtual;61-virtual;70-virtual;75-virtual'
+              -DCMAKE_CUDA_ARCHITECTURES='50-virtual;52-virtual;61-virtual;70-virtual;75-virtual'
             cmake --build build -j$(nproc)
             ccache -s
       - run:
@@ -1387,18 +1360,17 @@ jobs:
             - "*.whl"
 
   build-py-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
-    resource_class: macos.m1.large.gen1
     steps:
       - checkout
       - restore_cache:
           keys:
             - ccache-gpt4all-macos-
+      - run:
+          <<: *job-macos-install-deps
       - run:
           name: Install dependencies
           command: |
-            brew install ccache cmake
             pip install setuptools wheel cmake
       - run:
           name: Build C library
@@ -1409,6 +1381,9 @@ jobs:
             cd gpt4all-backend
             cmake -B build \
               -DCMAKE_BUILD_TYPE=Release \
+              -DCMAKE_C_COMPILER=/opt/homebrew/opt/llvm/bin/clang \
+              -DCMAKE_CXX_COMPILER=/opt/homebrew/opt/llvm/bin/clang++ \
+              -DCMAKE_RANLIB=/usr/bin/ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DBUILD_UNIVERSAL=ON \
@@ -1483,7 +1458,7 @@ jobs:
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache `
               -DCMAKE_CUDA_COMPILER_LAUNCHER=ccache `
               -DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON `
-              -DCMAKE_CUDA_ARCHITECTURES='52-virtual;61-virtual;70-virtual;75-virtual'
+              -DCMAKE_CUDA_ARCHITECTURES='50-virtual;52-virtual;61-virtual;70-virtual;75-virtual'
             cmake --build build --parallel
             ccache -s
       - run:
@@ -1537,18 +1512,7 @@ jobs:
           keys:
             - ccache-gpt4all-linux-amd64-
       - run:
-          name: Install dependencies
+          <<: *job-linux-install-backend-deps
-          command: |
-            wget -qO- "https://packages.lunarg.com/lunarg-signing-key-pub.asc" | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc
-            wget -qO- "https://packages.lunarg.com/vulkan/1.3.290/lunarg-vulkan-1.3.290-jammy.list" | sudo tee /etc/apt/sources.list.d/lunarg-vulkan-1.3.290-jammy.list
-            wget "https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb"
-            sudo dpkg -i cuda-keyring_1.1-1_all.deb
-            packages=(
-              build-essential ccache cmake cuda-compiler-11-8 g++-12 libcublas-dev-11-8 libnvidia-compute-550-server
-              vulkan-sdk
-            )
-            sudo apt-get update
-            sudo apt-get install -y "${packages[@]}"
       - run:
           name: Build Libraries
           no_output_timeout: 30m
@@ -1558,10 +1522,12 @@ jobs:
             cd gpt4all-backend
             mkdir -p runtimes/build
             cd runtimes/build
-            cmake ../.. \
+            cmake ../.. -G Ninja \
               -DCMAKE_BUILD_TYPE=Release \
-              -DCMAKE_C_COMPILER=gcc-12 \
+              -DCMAKE_C_COMPILER=clang-19 \
-              -DCMAKE_C_COMPILER=g++-12 \
+              -DCMAKE_CXX_COMPILER=clang++-19 \
+              -DCMAKE_CXX_COMPILER_AR=ar \
+              -DCMAKE_CXX_COMPILER_RANLIB=ranlib \
               -DCMAKE_BUILD_TYPE=Release \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
@@ -1582,8 +1548,7 @@ jobs:
             - runtimes/linux-x64/*.so
 
   build-bindings-backend-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - run:
@@ -1595,9 +1560,7 @@ jobs:
           keys:
             - ccache-gpt4all-macos-
       - run:
-          name: Install dependencies
+          <<: *job-macos-install-deps
-          command: |
-            brew install ccache cmake
       - run:
           name: Build Libraries
           no_output_timeout: 30m
@@ -1608,6 +1571,9 @@ jobs:
             cd runtimes/build
             cmake ../.. \
               -DCMAKE_BUILD_TYPE=Release \
+              -DCMAKE_C_COMPILER=/opt/homebrew/opt/llvm/bin/clang \
+              -DCMAKE_CXX_COMPILER=/opt/homebrew/opt/llvm/bin/clang++ \
+              -DCMAKE_RANLIB=/usr/bin/ranlib \
               -DCMAKE_C_COMPILER_LAUNCHER=ccache \
               -DCMAKE_CXX_COMPILER_LAUNCHER=ccache \
               -DBUILD_UNIVERSAL=ON \
@@ -1725,8 +1691,7 @@ jobs:
             - runtimes/linux-x64/*-*.so
 
   build-nodejs-macos:
-    macos:
+    <<: *job-macos-executor
-      xcode: 15.4.0
     steps:
       - checkout
       - attach_workspace:
@@ -1903,22 +1868,22 @@ workflows:
           context: gpt4all
       - build-offline-chat-installer-linux:
           context: gpt4all
-      - sign-online-chat-installer-macos:
+      - sign-offline-chat-installer-macos:
           context: gpt4all
           requires:
-            - build-online-chat-installer-macos
+            - build-offline-chat-installer-macos
-      - notarize-online-chat-installer-macos:
+      - notarize-offline-chat-installer-macos:
           context: gpt4all
           requires:
-            - sign-online-chat-installer-macos
+            - sign-offline-chat-installer-macos
-      - sign-online-chat-installer-windows:
+      - sign-offline-chat-installer-windows:
           context: gpt4all
           requires:
-            - build-online-chat-installer-windows
+            - build-offline-chat-installer-windows
-      - sign-online-chat-installer-windows-arm:
+      - sign-offline-chat-installer-windows-arm:
           context: gpt4all
           requires:
-            - build-online-chat-installer-windows-arm
+            - build-offline-chat-installer-windows-arm
   build-chat-installers-release:
     # only run on main branch tags that start with 'v' and a digit
     when:

MAINTAINERS.md

@@ -72,6 +72,6 @@ Discord: `@Tim453`
 - Flatpak
 
 Jack ([@wuodoo](https://github.com/wuodoo))<br/>
-E-mail: 2296103047@qq.com><br/>
+E-mail: 2296103047@qq.com<br/>
 Discord: `@mikage`
 - zh\_CN translation

README.md

@@ -35,6 +35,11 @@ GPT4All is made possible by our compute partner <a href="https://www.paperspace.
     <img src="gpt4all-bindings/python/docs/assets/windows.png" style="height: 1em; width: auto" /> Windows Installer
   </a> &mdash;
 </p>
+<p>
+  &mdash; <a href="https://gpt4all.io/installers/gpt4all-installer-win64-arm.exe">
+    <img src="gpt4all-bindings/python/docs/assets/windows.png" style="height: 1em; width: auto" /> Windows ARM Installer
+  </a> &mdash;
+</p>
 <p>
   &mdash; <a href="https://gpt4all.io/installers/gpt4all-installer-darwin.dmg">
     <img src="gpt4all-bindings/python/docs/assets/mac.png" style="height: 1em; width: auto" /> macOS Installer
@@ -46,10 +51,16 @@ GPT4All is made possible by our compute partner <a href="https://www.paperspace.
   </a> &mdash;
 </p>
 <p>
-  Windows and Linux require Intel Core i3 2nd Gen / AMD Bulldozer, or better. x86-64 only, no ARM.
+  The Windows and Linux builds require Intel Core i3 2nd Gen / AMD Bulldozer, or better.
 </p>
 <p>
-  macOS requires Monterey 12.6 or newer. Best results with Apple Silicon M-series processors.
+  The Windows ARM build supports Qualcomm Snapdragon and Microsoft SQ1/SQ2 processors.
+</p>
+<p>
+  The Linux build is x86-64 only (no ARM).
+</p>
+<p>
+  The macOS build requires Monterey 12.6 or newer. Best results with Apple Silicon M-series processors.
 </p>
 
 See the full [System Requirements](gpt4all-chat/system_requirements.md) for more details.

gpt4all-backend/CMakeLists.txt

@@ -69,7 +69,7 @@ if (LLMODEL_CUDA)
     cmake_minimum_required(VERSION 3.18)  # for CMAKE_CUDA_ARCHITECTURES
 
     # Defaults must be set before enable_language(CUDA).
-    # Keep this in sync with the arch list in ggml/src/CMakeLists.txt.
+    # Keep this in sync with the arch list in ggml/src/CMakeLists.txt (plus 5.0 for non-F16 branch).
     if (NOT DEFINED CMAKE_CUDA_ARCHITECTURES)
         # 52 == lowest CUDA 12 standard
         # 60 == f16 CUDA intrinsics
@@ -78,7 +78,7 @@ if (LLMODEL_CUDA)
         if (GGML_CUDA_F16 OR GGML_CUDA_DMMV_F16)
             set(CMAKE_CUDA_ARCHITECTURES "60;61;70;75") # needed for f16 CUDA intrinsics
         else()
-            set(CMAKE_CUDA_ARCHITECTURES "52;61;70;75") # lowest CUDA 12 standard + lowest for integer intrinsics
+            set(CMAKE_CUDA_ARCHITECTURES "50;52;61;70;75") # lowest CUDA 12 standard + lowest for integer intrinsics
             #set(CMAKE_CUDA_ARCHITECTURES "OFF") # use this to compile much faster, but only F16 models work
         endif()
     endif()

gpt4all-backend/deps/llama.cpp-mainline

@@ -1 +1 @@
-Subproject commit 3ebb3603e807d74a16f061c46d2925a1653e7a93
+Subproject commit 11f734c3b0334dbae4823b4a7467764e447fc6d6

gpt4all-backend/src/llamamodel.cpp

@@ -53,6 +53,7 @@ static const std::vector<const char *> KNOWN_ARCHES {
     "gpt2",
     // "gptj", -- no inference code
     "gptneox",
+    "granite",
     "granitemoe",
     "mpt",
     "baichuan",

gpt4all-backend/src/llmodel.cpp

@@ -140,9 +140,14 @@ const std::vector<LLModel::Implementation> &LLModel::Implementation::implementat
             std::string path;
             // Split the paths string by the delimiter and process each path.
             while (std::getline(ss, path, ';')) {
-                std::u8string u8_path(path.begin(), path.end());
+                fs::directory_iterator iter;
+                try {
+                    iter = fs::directory_iterator(std::u8string(path.begin(), path.end()));
+                } catch (const fs::filesystem_error &) {
+                    continue; // skip nonexistent path
+                }
                 // Iterate over all libraries
-                for (const auto &f : fs::directory_iterator(u8_path)) {
+                for (const auto &f : iter) {
                     const fs::path &p = f.path();
 
                     if (p.extension() != LIB_FILE_EXT) continue;

gpt4all-chat/CHANGELOG.md

@@ -4,13 +4,30 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
+## [3.10.0] - 2025-02-24
+
+### Added
+- Whitelist Granite (non-MoE) model architecture (by [@ThiloteE](https://github.com/ThiloteE) in [#3487](https://github.com/nomic-ai/gpt4all/pull/3487))
+- Add support for CUDA compute 5.0 GPUs such as the GTX 750 ([#3499](https://github.com/nomic-ai/gpt4all/pull/3499))
+- Add a Remote Providers tab to the Add Model page ([#3506](https://github.com/nomic-ai/gpt4all/pull/3506))
+
+### Changed
+- Substitute prettier default templates for OLMoE 7B 0924/0125 and Granite 3.1 3B/8B (by [@ThiloteE](https://github.com/ThiloteE) in [#3471](https://github.com/nomic-ai/gpt4all/pull/3471))
+- Build with LLVM Clang 19 on macOS and Ubuntu ([#3500](https://github.com/nomic-ai/gpt4all/pull/3500))
+
+### Fixed
+- Fix several potential crashes ([#3465](https://github.com/nomic-ai/gpt4all/pull/3465))
+- Fix visual spacing issues with deepseek models ([#3470](https://github.com/nomic-ai/gpt4all/pull/3470))
+- Add missing strings to Italian translation (by [@Harvester62](https://github.com/Harvester62) in [#3496](https://github.com/nomic-ai/gpt4all/pull/3496))
+- Update Simplified Chinese translation (by [@Junior2Ran](https://github.com/Junior2Ran) in [#3467](https://github.com/nomic-ai/pull/3467))
+
 ## [3.9.0] - 2025-02-04
 
 ### Added
 - Whitelist OLMoE and Granite MoE model architectures (no Vulkan) (by [@ThiloteE](https://github.com/ThiloteE) in [#3449](https://github.com/nomic-ai/gpt4all/pull/3449))
 
 ### Fixed
-- Fix "index N is not a prompt" when using LocalDocs with reasoning ([#3451](https://github.com/nomic-ai/gpt4all/pull/3451)
+- Fix "index N is not a prompt" when using LocalDocs with reasoning ([#3451](https://github.com/nomic-ai/gpt4all/pull/3451))
 - Work around rendering artifacts on Snapdragon SoCs with Windows ([#3450](https://github.com/nomic-ai/gpt4all/pull/3450))
 - Prevent DeepSeek-R1 reasoning from appearing in chat names and follow-up questions ([#3458](https://github.com/nomic-ai/gpt4all/pull/3458))
 - Fix LocalDocs crash on Windows ARM when reading PDFs ([#3460](https://github.com/nomic-ai/gpt4all/pull/3460))
@@ -295,6 +312,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 - Fix several Vulkan resource management issues ([#2694](https://github.com/nomic-ai/gpt4all/pull/2694))
 - Fix crash/hang when some models stop generating, by showing special tokens ([#2701](https://github.com/nomic-ai/gpt4all/pull/2701))
 
+[3.10.0]: https://github.com/nomic-ai/gpt4all/compare/v3.9.0...v3.10.0
 [3.9.0]: https://github.com/nomic-ai/gpt4all/compare/v3.8.0...v3.9.0
 [3.8.0]: https://github.com/nomic-ai/gpt4all/compare/v3.7.0...v3.8.0
 [3.7.0]: https://github.com/nomic-ai/gpt4all/compare/v3.6.1...v3.7.0

gpt4all-chat/CMakeLists.txt

@@ -3,10 +3,10 @@ cmake_minimum_required(VERSION 3.25)  # for try_compile SOURCE_FROM_VAR
 include(../common/common.cmake)
 
 set(APP_VERSION_MAJOR 3)
-set(APP_VERSION_MINOR 9)
+set(APP_VERSION_MINOR 10)
-set(APP_VERSION_PATCH 0)
+set(APP_VERSION_PATCH 1)
 set(APP_VERSION_BASE "${APP_VERSION_MAJOR}.${APP_VERSION_MINOR}.${APP_VERSION_PATCH}")
-set(APP_VERSION "${APP_VERSION_BASE}")
+set(APP_VERSION "${APP_VERSION_BASE}-dev0")
 
 project(gpt4all VERSION ${APP_VERSION_BASE} LANGUAGES CXX C)
 
@@ -104,7 +104,7 @@ elseif (GPT4ALL_USE_QTPDF MATCHES "^(ON|AUTO)$")
     set(GPT4ALL_USING_QTPDF ON)
     list(APPEND GPT4ALL_QT_COMPONENTS Pdf)
 endif()
-find_package(Qt6 6.5 COMPONENTS ${GPT4ALL_QT_COMPONENTS} REQUIRED)
+find_package(Qt6 6.8 COMPONENTS ${GPT4ALL_QT_COMPONENTS} REQUIRED)
 
 if (QT_KNOWN_POLICY_QTP0004)
     qt_policy(SET QTP0004 NEW)  # generate extra qmldir files on Qt 6.8+
@@ -266,6 +266,7 @@ qt_add_qml_module(chat
       qml/AddModelView.qml
       qml/AddGPT4AllModelView.qml
       qml/AddHFModelView.qml
+      qml/AddRemoteModelView.qml
       qml/ApplicationSettings.qml
       qml/ChatDrawer.qml
       qml/ChatCollapsibleItem.qml
@@ -314,6 +315,7 @@ qt_add_qml_module(chat
       qml/MyTextField.qml
       qml/MyToolButton.qml
       qml/MyWelcomeButton.qml
+      qml/RemoteModelCard.qml
     RESOURCES
       icons/antenna_1.svg
       icons/antenna_2.svg
@@ -344,6 +346,7 @@ qt_add_qml_module(chat
       icons/gpt4all-48.png
       icons/gpt4all.svg
       icons/gpt4all_transparent.svg
+      icons/groq.svg
       icons/home.svg
       icons/image.svg
       icons/info.svg
@@ -351,12 +354,14 @@ qt_add_qml_module(chat
       icons/left_panel_open.svg
       icons/local-docs.svg
       icons/models.svg
+      icons/mistral.svg
       icons/network.svg
       icons/nomic_logo.svg
       icons/notes.svg
       icons/paperclip.svg
       icons/plus.svg
       icons/plus_circle.svg
+      icons/openai.svg
       icons/recycle.svg
       icons/regenerate.svg
       icons/search.svg

gpt4all-chat/icons/groq.svg

@@ -0,0 +1,3 @@
+<?xml version="1.0" encoding="utf-8" ?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 26.3 26.3"><defs><style>.cls-1{fill:#f05237;}.cls-2{fill:#fff;}</style></defs><g id="Layer_2" data-name="Layer 2"><g id="Content"><circle class="cls-1" cx="13.15" cy="13.15" r="13.15"/><path class="cls-2" d="M13.17,6.88a4.43,4.43,0,0,0,0,8.85h1.45V14.07H13.17a2.77,2.77,0,1,1,2.77-2.76v4.07a2.74,2.74,0,0,1-4.67,2L10.1,18.51a4.37,4.37,0,0,0,3.07,1.29h.06a4.42,4.42,0,0,0,4.36-4.4V11.2a4.43,4.43,0,0,0-4.42-4.32"/></g></g></svg>

gpt4all-chat/icons/mistral.svg

@@ -0,0 +1 @@
+<svg viewBox="0 0 512 512" xmlns="http://www.w3.org/2000/svg" fill-rule="evenodd" clip-rule="evenodd" stroke-linejoin="round" stroke-miterlimit="2"><path d="M189.08 303.228H94.587l.044-94.446h94.497l-.048 94.446z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M283.528 397.674h-94.493l.044-94.446h94.496l-.047 94.446z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M283.575 303.228H189.08l.046-94.446h94.496l-.047 94.446z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M378.07 303.228h-94.495l.044-94.446h94.498l-.047 94.446zM189.128 208.779H94.633l.044-94.448h94.498l-.047 94.448zM378.115 208.779h-94.494l.045-94.448h94.496l-.047 94.448zM94.587 303.227H.093l.044-96.017h94.496l-.046 96.017z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M94.633 208.779H.138l.046-94.448H94.68l-.047 94.448z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M94.68 115.902H.185L.23 19.885h94.498l-.047 96.017zM472.657 114.331h-94.495l.044-94.446h94.497l-.046 94.446zM94.54 399.244H.046l.044-97.588h94.497l-.047 97.588z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M94.495 492.123H0l.044-94.446H94.54l-.045 94.446zM472.563 303.228H378.07l.044-94.446h94.496l-.047 94.446zM472.61 208.779h-94.495l.044-94.448h94.498l-.047 94.448z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M472.517 397.674h-94.494l.044-94.446h94.497l-.047 94.446z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M472.47 492.121h-94.493l.044-96.017h94.496l-.047 96.017z" fill="#1c1c1b" fill-rule="nonzero"/><path d="M228.375 303.22h-96.061l.046-94.446h96.067l-.052 94.446z" fill="#ff7000" fill-rule="nonzero"/><path d="M322.827 397.666h-94.495l.044-96.018h94.498l-.047 96.018z" fill="#ff4900" fill-rule="nonzero"/><path d="M324.444 303.22h-97.636l.046-94.446h97.638l-.048 94.446z" fill="#ff7000" fill-rule="nonzero"/><path d="M418.938 303.22h-96.064l.045-94.446h96.066l-.047 94.446z" fill="#ff7000" fill-rule="nonzero"/><path d="M228.423 208.77H132.36l.045-94.445h96.066l-.05 94.446zM418.985 208.77H322.92l.044-94.445h96.069l-.048 94.446z" fill="#ffa300" fill-rule="nonzero"/><path d="M133.883 304.79H39.392l.044-96.017h94.496l-.049 96.017z" fill="#ff7000" fill-rule="nonzero"/><path d="M133.929 208.77H39.437l.044-95.445h94.496l-.048 95.445z" fill="#ffa300" fill-rule="nonzero"/><path d="M133.976 114.325H39.484l.044-94.448h94.497l-.05 94.448zM511.954 115.325h-94.493l.044-95.448h94.497l-.048 95.448z" fill="#ffce00" fill-rule="nonzero"/><path d="M133.836 399.667H39.345l.044-96.447h94.496l-.049 96.447z" fill="#ff4900" fill-rule="nonzero"/><path d="M133.79 492.117H39.3l.044-94.448h94.496l-.049 94.448z" fill="#ff0107" fill-rule="nonzero"/><path d="M511.862 303.22h-94.495l.046-94.446h94.496l-.047 94.446z" fill="#ff7000" fill-rule="nonzero"/><path d="M511.907 208.77h-94.493l.044-94.445h94.496l-.047 94.446z" fill="#ffa300" fill-rule="nonzero"/><path d="M511.815 398.666h-94.493l.044-95.447h94.496l-.047 95.447z" fill="#ff4900" fill-rule="nonzero"/><path d="M511.77 492.117h-94.496l.046-94.448h94.496l-.047 94.448z" fill="#ff0107" fill-rule="nonzero"/></svg>

gpt4all-chat/icons/openai.svg

@@ -0,0 +1,2 @@
+<?xml version="1.0" encoding="utf-8"?><!-- Uploaded to: SVG Repo, www.svgrepo.com, Generator: SVG Repo Mixer Tools -->
+<svg fill="#000000" width="800px" height="800px" viewBox="0 0 24 24" role="img" xmlns="http://www.w3.org/2000/svg"><title>OpenAI icon</title><path d="M22.2819 9.8211a5.9847 5.9847 0 0 0-.5157-4.9108 6.0462 6.0462 0 0 0-6.5098-2.9A6.0651 6.0651 0 0 0 4.9807 4.1818a5.9847 5.9847 0 0 0-3.9977 2.9 6.0462 6.0462 0 0 0 .7427 7.0966 5.98 5.98 0 0 0 .511 4.9107 6.051 6.051 0 0 0 6.5146 2.9001A5.9847 5.9847 0 0 0 13.2599 24a6.0557 6.0557 0 0 0 5.7718-4.2058 5.9894 5.9894 0 0 0 3.9977-2.9001 6.0557 6.0557 0 0 0-.7475-7.0729zm-9.022 12.6081a4.4755 4.4755 0 0 1-2.8764-1.0408l.1419-.0804 4.7783-2.7582a.7948.7948 0 0 0 .3927-.6813v-6.7369l2.02 1.1686a.071.071 0 0 1 .038.052v5.5826a4.504 4.504 0 0 1-4.4945 4.4944zm-9.6607-4.1254a4.4708 4.4708 0 0 1-.5346-3.0137l.142.0852 4.783 2.7582a.7712.7712 0 0 0 .7806 0l5.8428-3.3685v2.3324a.0804.0804 0 0 1-.0332.0615L9.74 19.9502a4.4992 4.4992 0 0 1-6.1408-1.6464zM2.3408 7.8956a4.485 4.485 0 0 1 2.3655-1.9728V11.6a.7664.7664 0 0 0 .3879.6765l5.8144 3.3543-2.0201 1.1685a.0757.0757 0 0 1-.071 0l-4.8303-2.7865A4.504 4.504 0 0 1 2.3408 7.872zm16.5963 3.8558L13.1038 8.364 15.1192 7.2a.0757.0757 0 0 1 .071 0l4.8303 2.7913a4.4944 4.4944 0 0 1-.6765 8.1042v-5.6772a.79.79 0 0 0-.407-.667zm2.0107-3.0231l-.142-.0852-4.7735-2.7818a.7759.7759 0 0 0-.7854 0L9.409 9.2297V6.8974a.0662.0662 0 0 1 .0284-.0615l4.8303-2.7866a4.4992 4.4992 0 0 1 6.6802 4.66zM8.3065 12.863l-2.02-1.1638a.0804.0804 0 0 1-.038-.0567V6.0742a4.4992 4.4992 0 0 1 7.3757-3.4537l-.142.0805L8.704 5.459a.7948.7948 0 0 0-.3927.6813zm1.0976-2.3654l2.602-1.4998 2.6069 1.4998v2.9994l-2.5974 1.4997-2.6067-1.4997Z"/></svg>

gpt4all-chat/metadata/latestnews.md

@@ -1,16 +1,15 @@
 ## Latest News
 
-GPT4All v3.8.0 was released on January 30th. Changes include:
+GPT4All v3.10.0 was released on February 24th. Changes include:
 
-* **Native DeepSeek-R1-Distill Support:** GPT4All now has robust support for the DeepSeek-R1 family of distillations.
+* **Remote Models:**
-  * Several model variants are now available on the downloads page.
+  * The Add Model page now has a dedicated tab for remote model providers.
-  * Reasoning (wrapped in "think" tags) is displayed similarly to the Reasoner model.
+  * Groq, OpenAI, and Mistral remote models are now easier to configure.
-  * The DeepSeek-R1 Qwen pretokenizer is now supported, resolving the loading failure in previous versions.
+* **CUDA Compatibility:** GPUs with CUDA compute capability 5.0 such as the GTX 750 are now supported by the CUDA backend.
-  * The model is now configured with a GPT4All-compatible prompt template by default.
+* **New Model:** The non-MoE Granite model is now supported.
-* **Chat Templating Overhaul:** The template parser has been *completely* replaced with one that has much better compatibility with common models.
+* **Translation Updates:**
-* **Code Interpreter Fixes:**
+  * The Italian translation has been updated.
-  * An issue preventing the code interpreter from logging a single string in v3.7.0 has been fixed.
+  * The Simplified Chinese translation has been significantly improved.
-  * The UI no longer freezes while the code interpreter is running a computation.
+* **Better Chat Templates:** The default chat templates for OLMoE 7B 0924/0125 and Granite 3.1 3B/8B have been improved.
-* **Local Server Fixes:**
+* **Whitespace Fixes:** DeepSeek-R1-based models now have better whitespace behavior in their output.
-  * An issue preventing the server from using LocalDocs after the first request since v3.5.0 has been fixed.
+* **Crash Fixes:** Several issues that could potentially cause GPT4All to crash have been fixed.
-  * System messages are now correctly hidden from the message history.

gpt4all-chat/metadata/release.json

@@ -268,5 +268,15 @@
     "version": "3.8.0",
     "notes": "* **Native DeepSeek-R1-Distill Support:** GPT4All now has robust support for the DeepSeek-R1 family of distillations.\n  * Several model variants are now available on the downloads page.\n  * Reasoning (wrapped in \"think\" tags) is displayed similarly to the Reasoner model.\n  * The DeepSeek-R1 Qwen pretokenizer is now supported, resolving the loading failure in previous versions.\n  * The model is now configured with a GPT4All-compatible prompt template by default.\n* **Chat Templating Overhaul:** The template parser has been *completely* replaced with one that has much better compatibility with common models.\n* **Code Interpreter Fixes:**\n  * An issue preventing the code interpreter from logging a single string in v3.7.0 has been fixed.\n  * The UI no longer freezes while the code interpreter is running a computation.\n* **Local Server Fixes:**\n  * An issue preventing the server from using LocalDocs after the first request since v3.5.0 has been fixed.\n  * System messages are now correctly hidden from the message history.\n",
     "contributors": "* Jared Van Bortel (Nomic AI)\n* Adam Treat (Nomic AI)\n* ThiloteE (`@ThiloteE`)"
+  },
+  {
+    "version": "3.9.0",
+    "notes": "* **LocalDocs Fix:** LocalDocs no longer shows an error on later messages with reasoning models.\n* **DeepSeek Fix:** DeepSeek-R1 reasoning (in 'think' tags) no longer appears in chat names and follow-up questions.\n* **Windows ARM Improvements:**\n  * Graphical artifacts on some SoCs have been fixed.\n  * A crash when adding a collection of PDFs to LocalDocs has been fixed.\n* **Template Parser Fixes:** Chat templates containing an unclosed comment no longer freeze GPT4All.\n* **New Models:** OLMoE and Granite MoE models are now supported.\n",
+    "contributors": "* Jared Van Bortel (Nomic AI)\n* Adam Treat (Nomic AI)\n* ThiloteE (`@ThiloteE`)"
+  },
+  {
+    "version": "3.10.0",
+    "notes": "* **Remote Models:**\n  * The Add Model page now has a dedicated tab for remote model providers.\n  * Groq, OpenAI, and Mistral remote models are now easier to configure.\n* **CUDA Compatibility:** GPUs with CUDA compute capability 5.0 such as the GTX 750 are now supported by the CUDA backend.\n* **New Model:** The non-MoE Granite model is now supported.\n* **Translation Updates:**\n  * The Italian translation has been updated.\n  * The Simplified Chinese translation has been significantly improved.\n* **Better Chat Templates:** The default chat templates for OLMoE 7B 0924/0125 and Granite 3.1 3B/8B have been improved.\n* **Whitespace Fixes:** DeepSeek-R1-based models now have better whitespace behavior in their output.\n* **Crash Fixes:** Several issues that could potentially cause GPT4All to crash have been fixed.\n",
+    "contributors": "* Jared Van Bortel (Nomic AI)\n* Adam Treat (Nomic AI)\n* ThiloteE (`@ThiloteE`)\n* Lil Bob (`@Junior2Ran`)\n* Riccardo Giovanetti (`@Harvester62`)"
   }
 ]

gpt4all-chat/qml/AddGPT4AllModelView.qml

@@ -204,7 +204,7 @@ ColumnLayout {
                                     Layout.minimumWidth: 200
                                     Layout.fillWidth: true
                                     Layout.alignment: Qt.AlignTop | Qt.AlignHCenter
-                                    visible: !isOnline && !installed && !calcHash && downloadError === ""
+                                    visible: !installed && !calcHash && downloadError === ""
                                     Accessible.description: qsTr("Stop/restart/start the download")
                                     onClicked: {
                                         if (!isDownloading) {
@@ -230,52 +230,6 @@ ColumnLayout {
                                     }
                                 }
 
-                                MySettingsButton {
-                                    id: installButton
-                                    visible: !installed && isOnline
-                                    Layout.topMargin: 20
-                                    Layout.leftMargin: 20
-                                    Layout.minimumWidth: 200
-                                    Layout.fillWidth: true
-                                    Layout.alignment: Qt.AlignTop | Qt.AlignHCenter
-                                    text: qsTr("Install")
-                                    font.pixelSize: theme.fontSizeLarge
-                                    onClicked: {
-                                        var apiKeyText = apiKey.text.trim(),
-                                        baseUrlText = baseUrl.text.trim(),
-                                        modelNameText = modelName.text.trim();
-
-                                        var apiKeyOk = apiKeyText !== "",
-                                        baseUrlOk = !isCompatibleApi || baseUrlText !== "",
-                                        modelNameOk = !isCompatibleApi || modelNameText !== "";
-
-                                        if (!apiKeyOk)
-                                            apiKey.showError();
-                                        if (!baseUrlOk)
-                                            baseUrl.showError();
-                                        if (!modelNameOk)
-                                            modelName.showError();
-
-                                        if (!apiKeyOk || !baseUrlOk || !modelNameOk)
-                                            return;
-
-                                        if (!isCompatibleApi)
-                                            Download.installModel(
-                                                        filename,
-                                                        apiKeyText,
-                                                        );
-                                        else
-                                            Download.installCompatibleModel(
-                                                        modelNameText,
-                                                        apiKeyText,
-                                                        baseUrlText,
-                                                        );
-                                    }
-                                    Accessible.role: Accessible.Button
-                                    Accessible.name: qsTr("Install")
-                                    Accessible.description: qsTr("Install online model")
-                                }
-
                                 ColumnLayout {
                                     spacing: 0
                                     Label {
@@ -390,69 +344,6 @@ ColumnLayout {
                                         Accessible.description: qsTr("Displayed when the file hash is being calculated")
                                     }
                                 }
-
-                                MyTextField {
-                                    id: apiKey
-                                    visible: !installed && isOnline
-                                    Layout.topMargin: 20
-                                    Layout.leftMargin: 20
-                                    Layout.minimumWidth: 200
-                                    Layout.alignment: Qt.AlignTop | Qt.AlignHCenter
-                                    wrapMode: Text.WrapAnywhere
-                                    function showError() {
-                                        messageToast.show(qsTr("ERROR: $API_KEY is empty."));
-                                        apiKey.placeholderTextColor = theme.textErrorColor;
-                                    }
-                                    onTextChanged: {
-                                        apiKey.placeholderTextColor = theme.mutedTextColor;
-                                    }
-                                    placeholderText: qsTr("enter $API_KEY")
-                                    Accessible.role: Accessible.EditableText
-                                    Accessible.name: placeholderText
-                                    Accessible.description: qsTr("Whether the file hash is being calculated")
-                                }
-
-                                MyTextField {
-                                    id: baseUrl
-                                    visible: !installed && isOnline && isCompatibleApi
-                                    Layout.topMargin: 20
-                                    Layout.leftMargin: 20
-                                    Layout.minimumWidth: 200
-                                    Layout.alignment: Qt.AlignTop | Qt.AlignHCenter
-                                    wrapMode: Text.WrapAnywhere
-                                    function showError() {
-                                        messageToast.show(qsTr("ERROR: $BASE_URL is empty."));
-                                        baseUrl.placeholderTextColor = theme.textErrorColor;
-                                    }
-                                    onTextChanged: {
-                                        baseUrl.placeholderTextColor = theme.mutedTextColor;
-                                    }
-                                    placeholderText: qsTr("enter $BASE_URL")
-                                    Accessible.role: Accessible.EditableText
-                                    Accessible.name: placeholderText
-                                    Accessible.description: qsTr("Whether the file hash is being calculated")
-                                }
-
-                                MyTextField {
-                                    id: modelName
-                                    visible: !installed && isOnline && isCompatibleApi
-                                    Layout.topMargin: 20
-                                    Layout.leftMargin: 20
-                                    Layout.minimumWidth: 200
-                                    Layout.alignment: Qt.AlignTop | Qt.AlignHCenter
-                                    wrapMode: Text.WrapAnywhere
-                                    function showError() {
-                                        messageToast.show(qsTr("ERROR: $MODEL_NAME is empty."))
-                                        modelName.placeholderTextColor = theme.textErrorColor;
-                                    }
-                                    onTextChanged: {
-                                        modelName.placeholderTextColor = theme.mutedTextColor;
-                                    }
-                                    placeholderText: qsTr("enter $MODEL_NAME")
-                                    Accessible.role: Accessible.EditableText
-                                    Accessible.name: placeholderText
-                                    Accessible.description: qsTr("Whether the file hash is being calculated")
-                                }
                             }
                         }
                     }

gpt4all-chat/qml/AddModelView.qml

@@ -89,6 +89,13 @@ Rectangle {
                     gpt4AllModelView.show();
                 }
             }
+            MyTabButton {
+                text: qsTr("Remote Providers")
+                isSelected: remoteModelView.isShown()
+                onPressed: {
+                    remoteModelView.show();
+                }
+            }
             MyTabButton {
                 text: qsTr("HuggingFace")
                 isSelected: huggingfaceModelView.isShown()
@@ -112,7 +119,20 @@ Rectangle {
                     stackLayout.currentIndex = 0;
                 }
                 function isShown() {
-                    return stackLayout.currentIndex === 0
+                    return stackLayout.currentIndex === 0;
+                }
+            }
+
+            AddRemoteModelView {
+                id: remoteModelView
+                Layout.fillWidth: true
+                Layout.fillHeight: true
+
+                function show() {
+                    stackLayout.currentIndex = 1;
+                }
+                function isShown() {
+                    return stackLayout.currentIndex === 1;
                 }
             }
 
@@ -126,10 +146,10 @@ Rectangle {
                 anchors.fill: parent
 
                 function show() {
-                    stackLayout.currentIndex = 1;
+                    stackLayout.currentIndex = 2;
                 }
                 function isShown() {
-                    return stackLayout.currentIndex === 1
+                    return stackLayout.currentIndex === 2;
                 }
             }
         }

gpt4all-chat/qml/AddRemoteModelView.qml

@@ -0,0 +1,147 @@
+import QtCore
+import QtQuick
+import QtQuick.Controls
+import QtQuick.Controls.Basic
+import QtQuick.Layouts
+import QtQuick.Dialogs
+import Qt.labs.folderlistmodel
+import Qt5Compat.GraphicalEffects
+
+import llm
+import chatlistmodel
+import download
+import modellist
+import network
+import gpt4all
+import mysettings
+import localdocs
+
+ColumnLayout {
+    Layout.fillWidth: true
+    Layout.alignment: Qt.AlignTop
+    spacing: 5
+
+    Label {
+        Layout.topMargin: 0
+        Layout.bottomMargin: 25
+        Layout.rightMargin: 150 * theme.fontScale
+        Layout.alignment: Qt.AlignTop
+        Layout.fillWidth: true
+        verticalAlignment: Text.AlignTop
+        text: qsTr("Various remote model providers that use network resources for inference.")
+        font.pixelSize: theme.fontSizeLarger
+        color: theme.textColor
+        wrapMode: Text.WordWrap
+    }
+
+    ScrollView {
+        id: scrollView
+        ScrollBar.vertical.policy: ScrollBar.AsNeeded
+        Layout.fillWidth: true
+        Layout.fillHeight: true
+        contentWidth: availableWidth
+        clip: true
+        Flow {
+            anchors.left: parent.left
+            anchors.right: parent.right
+            spacing: 20
+            bottomPadding: 20
+            property int childWidth: 330 * theme.fontScale
+            property int childHeight: 400 + 166 * theme.fontScale
+            RemoteModelCard {
+                width: parent.childWidth
+                height: parent.childHeight
+                providerBaseUrl: "https://api.groq.com/openai/v1/"
+                providerName: qsTr("Groq")
+                providerImage: "qrc:/gpt4all/icons/groq.svg"
+                providerDesc: qsTr('Groq offers a high-performance AI inference engine designed for low-latency and efficient processing. Optimized for real-time applications, Groq’s technology is ideal for users who need fast responses from open large language models and other AI workloads.<br><br>Get your API key: <a href="https://console.groq.com/keys">https://groq.com/</a>')
+                modelWhitelist: [
+                    // last updated 2025-02-24
+                    "deepseek-r1-distill-llama-70b",
+                    "deepseek-r1-distill-qwen-32b",
+                    "gemma2-9b-it",
+                    "llama-3.1-8b-instant",
+                    "llama-3.2-1b-preview",
+                    "llama-3.2-3b-preview",
+                    "llama-3.3-70b-specdec",
+                    "llama-3.3-70b-versatile",
+                    "llama3-70b-8192",
+                    "llama3-8b-8192",
+                    "mixtral-8x7b-32768",
+                    "qwen-2.5-32b",
+                    "qwen-2.5-coder-32b",
+                ]
+            }
+            RemoteModelCard {
+                width: parent.childWidth
+                height: parent.childHeight
+                providerBaseUrl: "https://api.openai.com/v1/"
+                providerName: qsTr("OpenAI")
+                providerImage: "qrc:/gpt4all/icons/openai.svg"
+                providerDesc: qsTr('OpenAI provides access to advanced AI models, including GPT-4 supporting a wide range of applications, from conversational AI to content generation and code completion.<br><br>Get your API key: <a href="https://platform.openai.com/signup">https://openai.com/</a>')
+                modelWhitelist: [
+                    // last updated 2025-02-24
+                    "gpt-3.5-turbo",
+                    "gpt-3.5-turbo-16k",
+                    "gpt-4",
+                    "gpt-4-32k",
+                    "gpt-4-turbo",
+                    "gpt-4o",
+                ]
+            }
+            RemoteModelCard {
+                width: parent.childWidth
+                height: parent.childHeight
+                providerBaseUrl: "https://api.mistral.ai/v1/"
+                providerName: qsTr("Mistral")
+                providerImage: "qrc:/gpt4all/icons/mistral.svg"
+                providerDesc: qsTr('Mistral AI specializes in efficient, open-weight language models optimized for various natural language processing tasks. Their models are designed for flexibility and performance, making them a solid option for applications requiring scalable AI solutions.<br><br>Get your API key: <a href="https://mistral.ai/">https://mistral.ai/</a>')
+                modelWhitelist: [
+                    // last updated 2025-02-24
+                    "codestral-2405",
+                    "codestral-2411-rc5",
+                    "codestral-2412",
+                    "codestral-2501",
+                    "codestral-latest",
+                    "codestral-mamba-2407",
+                    "codestral-mamba-latest",
+                    "ministral-3b-2410",
+                    "ministral-3b-latest",
+                    "ministral-8b-2410",
+                    "ministral-8b-latest",
+                    "mistral-large-2402",
+                    "mistral-large-2407",
+                    "mistral-large-2411",
+                    "mistral-large-latest",
+                    "mistral-medium-2312",
+                    "mistral-medium-latest",
+                    "mistral-saba-2502",
+                    "mistral-saba-latest",
+                    "mistral-small-2312",
+                    "mistral-small-2402",
+                    "mistral-small-2409",
+                    "mistral-small-2501",
+                    "mistral-small-latest",
+                    "mistral-tiny-2312",
+                    "mistral-tiny-2407",
+                    "mistral-tiny-latest",
+                    "open-codestral-mamba",
+                    "open-mistral-7b",
+                    "open-mistral-nemo",
+                    "open-mistral-nemo-2407",
+                    "open-mixtral-8x22b",
+                    "open-mixtral-8x22b-2404",
+                    "open-mixtral-8x7b",
+                ]
+            }
+            RemoteModelCard {
+                width: parent.childWidth
+                height: parent.childHeight
+                providerIsCustom: true
+                providerName: qsTr("Custom")
+                providerImage: "qrc:/gpt4all/icons/antenna_3.svg"
+                providerDesc: qsTr("The custom provider option allows users to connect their own OpenAI-compatible AI models or third-party inference services. This is useful for organizations with proprietary models or those leveraging niche AI providers not listed here.")
+            }
+        }
+    }
+}

gpt4all-chat/qml/ChatItemView.qml

@@ -198,6 +198,7 @@ GridLayout {
                         isError: false
                         isThinking: true
                         thinkingTime: modelData.thinkingTime
+                        visible: modelData.content !== ""
                     }
                 }
             }

gpt4all-chat/qml/MyComboBox.qml

@@ -60,27 +60,28 @@ ComboBox {
         highlighted: comboBox.highlightedIndex === index
     }
     popup: Popup {
-        // FIXME This should be made much nicer to take into account lists that are very long so
-        // that it is scrollable and also sized optimally taking into account the x,y and the content
-        // width and height as well as the window width and height
         y: comboBox.height - 1
         width: comboBox.width
-        implicitHeight: contentItem.implicitHeight + 20
+        implicitHeight: Math.min(window.height - y, contentItem.implicitHeight + 20)
         padding: 0
-
         contentItem: Rectangle {
-            implicitWidth: myListView.contentWidth
+            implicitWidth: comboBox.width
             implicitHeight: myListView.contentHeight
             color: "transparent"
-            ListView {
+            radius: 10
-                id: myListView
+            ScrollView {
                 anchors.fill: parent
                 anchors.margins: 10
                 clip: true
-                implicitHeight: contentHeight
+                ScrollBar.vertical.policy: ScrollBar.AsNeeded
-                model: comboBox.popup.visible ? comboBox.delegateModel : null
+                ScrollBar.horizontal.policy: ScrollBar.AlwaysOff
-                currentIndex: comboBox.highlightedIndex
+                ListView {
-                ScrollIndicator.vertical: ScrollIndicator { }
+                    id: myListView
+                    implicitHeight: contentHeight
+                    model: comboBox.popup.visible ? comboBox.delegateModel : null
+                    currentIndex: comboBox.highlightedIndex
+                    ScrollIndicator.vertical: ScrollIndicator { }
+                }
             }
         }
 

gpt4all-chat/qml/RemoteModelCard.qml

@@ -0,0 +1,221 @@
+import QtCore
+import QtQuick
+import QtQuick.Controls
+import QtQuick.Controls.Basic
+import QtQuick.Layouts
+import QtQuick.Dialogs
+import Qt.labs.folderlistmodel
+import Qt5Compat.GraphicalEffects
+
+import llm
+import chatlistmodel
+import download
+import modellist
+import network
+import gpt4all
+import mysettings
+import localdocs
+
+
+Rectangle {
+    property alias providerName: providerNameLabel.text
+    property alias providerImage: myimage.source
+    property alias providerDesc: providerDescLabel.text
+    property string providerBaseUrl: ""
+    property bool providerIsCustom: false
+    property var modelWhitelist: null
+
+    color: theme.conversationBackground
+    radius: 10
+    border.width: 1
+    border.color: theme.controlBorder
+    implicitHeight: topColumn.height + bottomColumn.height + 33 * theme.fontScale
+
+    ColumnLayout {
+        id: topColumn
+        anchors.left: parent.left
+        anchors.right: parent.right
+        anchors.top: parent.top
+        anchors.margins: 20
+        spacing: 15 * theme.fontScale
+        RowLayout {
+            Layout.alignment: Qt.AlignTop
+            spacing: 10
+            Item {
+                Layout.preferredWidth: 27 * theme.fontScale
+                Layout.preferredHeight: 27 * theme.fontScale
+                Layout.alignment: Qt.AlignLeft
+
+                Image {
+                    id: myimage
+                    anchors.centerIn: parent
+                    sourceSize.width: parent.width
+                    sourceSize.height: parent.height
+                    mipmap: true
+                    fillMode: Image.PreserveAspectFit
+                }
+            }
+
+            Label {
+                id: providerNameLabel
+                color: theme.textColor
+                font.pixelSize: theme.fontSizeBanner
+            }
+        }
+
+        Label {
+            id: providerDescLabel
+            Layout.fillWidth: true
+            wrapMode: Text.Wrap
+            color: theme.settingsTitleTextColor
+            font.pixelSize: theme.fontSizeLarge
+            onLinkActivated: function(link) { Qt.openUrlExternally(link); }
+
+            MouseArea {
+                anchors.fill: parent
+                acceptedButtons: Qt.NoButton // pass clicks to parent
+                cursorShape: parent.hoveredLink ? Qt.PointingHandCursor : Qt.ArrowCursor
+            }
+        }
+    }
+
+    ColumnLayout {
+        id: bottomColumn
+        anchors.left: parent.left
+        anchors.right: parent.right
+        anchors.bottom: parent.bottom
+        anchors.margins: 20
+        spacing: 30
+
+        ColumnLayout {
+            MySettingsLabel {
+                text: qsTr("API Key")
+                font.bold: true
+                font.pixelSize: theme.fontSizeLarge
+                color: theme.settingsTitleTextColor
+            }
+
+            MyTextField {
+                id: apiKeyField
+                Layout.fillWidth: true
+                font.pixelSize: theme.fontSizeLarge
+                wrapMode: Text.WrapAnywhere
+                function showError() {
+                    messageToast.show(qsTr("ERROR: $API_KEY is empty."));
+                    apiKeyField.placeholderTextColor = theme.textErrorColor;
+                }
+                onTextChanged: {
+                    apiKeyField.placeholderTextColor = theme.mutedTextColor;
+                    if (!providerIsCustom) {
+                        let models = ModelList.remoteModelList(apiKeyField.text, providerBaseUrl);
+                        if (modelWhitelist !== null)
+                            models = models.filter(m => modelWhitelist.includes(m));
+                        myModelList.model = models;
+                        myModelList.currentIndex = -1;
+                    }
+                }
+                placeholderText: qsTr("enter $API_KEY")
+                Accessible.role: Accessible.EditableText
+                Accessible.name: placeholderText
+                Accessible.description: qsTr("Whether the file hash is being calculated")
+            }
+        }
+
+        ColumnLayout {
+            visible: providerIsCustom
+            MySettingsLabel {
+                text: qsTr("Base Url")
+                font.bold: true
+                font.pixelSize: theme.fontSizeLarge
+                color: theme.settingsTitleTextColor
+            }
+            MyTextField {
+                id: baseUrlField
+                Layout.fillWidth: true
+                font.pixelSize: theme.fontSizeLarge
+                wrapMode: Text.WrapAnywhere
+                function showError() {
+                    messageToast.show(qsTr("ERROR: $BASE_URL is empty."));
+                    baseUrlField.placeholderTextColor = theme.textErrorColor;
+                }
+                onTextChanged: {
+                    baseUrlField.placeholderTextColor = theme.mutedTextColor;
+                }
+                placeholderText: qsTr("enter $BASE_URL")
+                Accessible.role: Accessible.EditableText
+                Accessible.name: placeholderText
+            }
+        }
+        ColumnLayout {
+            visible: providerIsCustom
+            MySettingsLabel {
+                text: qsTr("Model Name")
+                font.bold: true
+                font.pixelSize: theme.fontSizeLarge
+                color: theme.settingsTitleTextColor
+            }
+            MyTextField {
+                id: modelNameField
+                Layout.fillWidth: true
+                font.pixelSize: theme.fontSizeLarge
+                wrapMode: Text.WrapAnywhere
+                function showError() {
+                    messageToast.show(qsTr("ERROR: $MODEL_NAME is empty."))
+                    modelNameField.placeholderTextColor = theme.textErrorColor;
+                }
+                onTextChanged: {
+                    modelNameField.placeholderTextColor = theme.mutedTextColor;
+                }
+                placeholderText: qsTr("enter $MODEL_NAME")
+                Accessible.role: Accessible.EditableText
+                Accessible.name: placeholderText
+            }
+        }
+
+        ColumnLayout {
+            visible: myModelList.count > 0 && !providerIsCustom
+
+            MySettingsLabel {
+                text: qsTr("Models")
+                font.bold: true
+                font.pixelSize: theme.fontSizeLarge
+                color: theme.settingsTitleTextColor
+            }
+
+            RowLayout {
+                spacing: 10
+
+                MyComboBox {
+                    Layout.fillWidth: true
+                    id: myModelList
+                    currentIndex: -1;
+                }
+            }
+        }
+
+        MySettingsButton {
+            id: installButton
+            Layout.alignment: Qt.AlignRight
+            text: qsTr("Install")
+            font.pixelSize: theme.fontSizeLarge
+
+            property string apiKeyText: apiKeyField.text.trim()
+            property string baseUrlText: providerIsCustom ? baseUrlField.text.trim() : providerBaseUrl.trim()
+            property string modelNameText: providerIsCustom ? modelNameField.text.trim() : myModelList.currentText.trim()
+
+            enabled: apiKeyText !== "" && baseUrlText !== "" && modelNameText !== ""
+
+            onClicked: {
+                Download.installCompatibleModel(
+                            modelNameText,
+                            apiKeyText,
+                            baseUrlText,
+                            );
+                myModelList.currentIndex = -1;
+            }
+            Accessible.role: Accessible.Button
+            Accessible.name: qsTr("Install")
+            Accessible.description: qsTr("Install remote model")
+        }
+    }
+}

gpt4all-chat/src/chat.cpp

@@ -7,24 +7,26 @@
 #include "toolcallparser.h"
 #include "toolmodel.h"
 
-#include <QBuffer>
+#include <QByteArray>
 #include <QDataStream>
 #include <QDebug>
-#include <QJsonDocument>
+#include <QFile>
-#include <QJsonObject>
+#include <QFileInfo>
-#include <QJsonValue>
+#include <QIODevice>
 #include <QLatin1String>
 #include <QMap>
 #include <QRegularExpression>
 #include <QString>
-#include <QVariant>
 #include <Qt>
+#include <QtAssert>
 #include <QtLogging>
 
+#include <optional>
 #include <utility>
 
 using namespace ToolEnums;
 
+
 Chat::Chat(QObject *parent)
     : QObject(parent)
     , m_id(Network::globalInstance()->generateUniqueId())

gpt4all-chat/src/chat.h

@@ -3,21 +3,26 @@
 
 #include "chatllm.h"
 #include "chatmodel.h"
-#include "database.h" // IWYU pragma: keep
+#include "database.h"
-#include "localdocsmodel.h" // IWYU pragma: keep
+#include "localdocsmodel.h"
 #include "modellist.h"
+#include "tool.h"
 
 #include <QDateTime>
 #include <QList>
 #include <QObject>
-#include <QQmlEngine>
+#include <QQmlEngine> // IWYU pragma: keep
 #include <QString>
 #include <QStringList> // IWYU pragma: keep
-#include <QStringView>
+#include <QUrl>
-#include <QtGlobal>
+#include <QVariant>
+#include <QtTypes>
 
+// IWYU pragma: no_forward_declare LocalDocsCollectionsModel
+// IWYU pragma: no_forward_declare ToolCallInfo
 class QDataStream;
 
+
 class Chat : public QObject
 {
     Q_OBJECT

gpt4all-chat/src/chatapi.cpp

@@ -2,6 +2,9 @@
 
 #include "utils.h"
 
+#include <fmt/format.h>
+
+#include <QAnyStringView>
 #include <QCoreApplication>
 #include <QDebug>
 #include <QGuiApplication>
@@ -9,15 +12,17 @@
 #include <QJsonDocument>
 #include <QJsonObject>
 #include <QJsonValue>
+#include <QLatin1String>
 #include <QNetworkAccessManager>
 #include <QNetworkRequest>
+#include <QStringView>
 #include <QThread>
 #include <QUrl>
-#include <QUtf8StringView>
+#include <QUtf8StringView> // IWYU pragma: keep
 #include <QVariant>
 #include <QXmlStreamReader>
 #include <Qt>
-#include <QtGlobal>
+#include <QtAssert>
 #include <QtLogging>
 
 #include <expected>
@@ -29,6 +34,7 @@ using namespace Qt::Literals::StringLiterals;
 
 //#define DEBUG
 
+
 ChatAPI::ChatAPI()
     : QObject(nullptr)
     , m_modelName("gpt-3.5-turbo")

gpt4all-chat/src/chatapi.h

@@ -3,10 +3,11 @@
 
 #include <gpt4all-backend/llmodel.h>
 
-#include <QByteArray> // IWYU pragma: keep
+#include <QByteArray>
 #include <QNetworkReply>
 #include <QObject>
 #include <QString>
+#include <QtPreprocessorSupport>
 
 #include <cstddef>
 #include <cstdint>
@@ -17,9 +18,11 @@
 #include <unordered_map>
 #include <vector>
 
+// IWYU pragma: no_forward_declare QByteArray
+class ChatAPI;
 class QNetworkAccessManager;
 
-class ChatAPI;
+
 class ChatAPIWorker : public QObject {
     Q_OBJECT
 public:

gpt4all-chat/src/chatlistmodel.cpp

@@ -1,23 +1,24 @@
 #include "chatlistmodel.h"
 
-#include "database.h" // IWYU pragma: keep
 #include "mysettings.h"
 
+#include <QCoreApplication>
 #include <QDataStream>
 #include <QDir>
 #include <QElapsedTimer>
+#include <QEvent>
 #include <QFile>
 #include <QFileInfo>
 #include <QGlobalStatic>
 #include <QGuiApplication>
 #include <QIODevice>
 #include <QSettings>
-#include <QString>
+#include <QStringList> // IWYU pragma: keep
-#include <QStringList>
 #include <Qt>
+#include <QtTypes>
 
 #include <algorithm>
-#include <memory>
+
 
 static constexpr quint32 CHAT_FORMAT_MAGIC   = 0xF5D553CC;
 static constexpr qint32  CHAT_FORMAT_VERSION = 12;

gpt4all-chat/src/chatlistmodel.h

@@ -7,17 +7,20 @@
 
 #include <QAbstractListModel>
 #include <QByteArray>
+#include <QDate>
 #include <QDebug>
 #include <QHash>
 #include <QList>
 #include <QMutex>
 #include <QObject>
+#include <QString>
 #include <QThread>
 #include <QVariant>
-#include <QVector>
+#include <QVector> // IWYU pragma: keep
 #include <Qt>
-#include <QtGlobal>
+#include <QtAssert>
 #include <QtLogging>
+#include <QtPreprocessorSupport>
 
 #include <memory>
 

gpt4all-chat/src/chatllm.cpp

@@ -15,32 +15,40 @@
 #include <minja/minja.hpp>
 #include <nlohmann/json.hpp>
 
+#include <QChar>
 #include <QDataStream>
 #include <QDebug>
 #include <QFile>
 #include <QGlobalStatic>
-#include <QIODevice>
+#include <QIODevice> // IWYU pragma: keep
 #include <QJsonDocument>
 #include <QJsonObject>
 #include <QJsonValue>
 #include <QMap>
-#include <QMutex>
+#include <QMutex> // IWYU pragma: keep
 #include <QMutexLocker> // IWYU pragma: keep
-#include <QRegularExpression>
+#include <QRegularExpression> // IWYU pragma: keep
-#include <QRegularExpressionMatch>
+#include <QRegularExpressionMatch> // IWYU pragma: keep
 #include <QSet>
+#include <QStringView>
 #include <QTextStream>
 #include <QUrl>
+#include <QVariant>
 #include <QWaitCondition>
 #include <Qt>
+#include <QtAssert>
 #include <QtLogging>
+#include <QtTypes> // IWYU pragma: keep
 
 #include <algorithm>
 #include <chrono>
 #include <cmath>
+#include <concepts>
 #include <cstddef>
+#include <cstdint>
 #include <ctime>
 #include <exception>
+#include <functional>
 #include <iomanip>
 #include <limits>
 #include <optional>
@@ -976,7 +984,8 @@ public:
     {
         Q_UNUSED(bufferIdx)
         try {
-            m_cllm->m_chatModel->setResponseValue(response);
+            QString r = response;
+            m_cllm->m_chatModel->setResponseValue(removeLeadingWhitespace(r));
         } catch (const std::exception &e) {
             // We have a try/catch here because the main thread might have removed the response from
             // the chatmodel by erasing the conversation during the response... the main thread sets
@@ -991,7 +1000,7 @@ public:
     bool onRegularResponse() override
     {
         auto respStr = QString::fromUtf8(m_result->response);
-        return onBufferResponse(removeLeadingWhitespace(respStr), 0);
+        return onBufferResponse(respStr, 0);
     }
 
     bool getStopGenerating() const override
@@ -1078,7 +1087,7 @@ auto ChatLLM::promptInternal(
     auto respStr = QString::fromUtf8(result.response);
     if (!respStr.isEmpty() && (std::as_const(respStr).back().isSpace() || finalBuffers.size() > 1)) {
         if (finalBuffers.size() > 1)
-            m_chatModel->setResponseValue(finalBuffers.last());
+            m_chatModel->setResponseValue(finalBuffers.last().trimmed());
         else
             m_chatModel->setResponseValue(respStr.trimmed());
         emit responseChanged();

gpt4all-chat/src/chatllm.h

@@ -2,7 +2,7 @@
 #define CHATLLM_H
 
 #include "chatmodel.h"
-#include "database.h" // IWYU pragma: keep
+#include "database.h"
 #include "modellist.h"
 
 #include <gpt4all-backend/llmodel.h>
@@ -10,29 +10,30 @@
 #include <QByteArray>
 #include <QElapsedTimer>
 #include <QFileInfo>
-#include <QList>      // IWYU pragma: keep
+#include <QList>
 #include <QObject>
 #include <QPointer>
 #include <QString>
 #include <QStringList> // IWYU pragma: keep
-#include <QStringView>
 #include <QThread>
 #include <QVariantMap> // IWYU pragma: keep
-#include <QtGlobal>
+#include <QtNumeric>
 
 #include <atomic>
-#include <cstdint>
 #include <memory>
 #include <optional>
 #include <span>
 #include <string>
+#include <string_view>
 #include <variant>
+#include <vector>
 
 using namespace Qt::Literals::StringLiterals;
 
-class ChatViewResponseHandler;
+class ChatLLM;
 class QDataStream;
 
+
 // NOTE: values serialized to disk, do not change or reuse
 enum class LLModelTypeV0 { // chat versions 2-5
     MPT       = 0,
@@ -89,9 +90,6 @@ inline LLModelTypeV1 parseLLModelTypeV0(int v0)
     }
 }
 
-class ChatLLM;
-class ChatModel;
-
 struct LLModelInfo {
     std::unique_ptr<LLModel> model;
     QFileInfo fileInfo;

gpt4all-chat/src/chatmodel.cpp

@@ -2,9 +2,11 @@
 
 #include <QDebug>
 #include <QMap>
-#include <QtGlobal>
+#include <QTextStream>
 #include <QtLogging>
 
+#include <exception>
+
 
 QList<ResultInfo> ChatItem::consolidateSources(const QList<ResultInfo> &sources)
 {

gpt4all-chat/src/chatmodel.h

@@ -4,32 +4,41 @@
 #include "database.h"
 #include "tool.h"
 #include "toolcallparser.h"
-#include "utils.h"
+#include "utils.h" // IWYU pragma: keep
 #include "xlsxtomd.h"
 
 #include <fmt/format.h>
 
-#include <QApplication>
 #include <QAbstractListModel>
 #include <QBuffer>
 #include <QByteArray>
 #include <QClipboard>
 #include <QDataStream>
-#include <QJsonDocument>
+#include <QFileInfo>
+#include <QGuiApplication>
+#include <QIODevice>
 #include <QHash>
 #include <QList>
+#include <QMutex>
+#include <QMutexLocker> // IWYU pragma: keep
 #include <QObject>
-#include <QPair>
+#include <QPair> // IWYU pragma: keep
 #include <QString>
+#include <QStringList> // IWYU pragma: keep
+#include <QUrl>
 #include <QVariant>
-#include <QVector>
 #include <Qt>
-#include <QtGlobal>
+#include <QtAssert>
+#include <QtPreprocessorSupport>
+#include <QtTypes>
 
 #include <algorithm>
 #include <iterator>
+#include <list>
+#include <optional>
 #include <ranges>
 #include <span>
+#include <stdexcept>
 #include <utility>
 #include <vector>
 
@@ -204,7 +213,8 @@ public:
         : QObject(nullptr)
     {
         moveToThread(parent->thread());
-        setParent(parent);
+        // setParent must be called from the thread the object lives in
+        QMetaObject::invokeMethod(this, [this, parent]() { this->setParent(parent); });
     }
 
     // NOTE: System messages are currently never serialized and only *stored* by the local server.

gpt4all-chat/src/chatviewtextprocessor.cpp

@@ -1,29 +1,32 @@
 #include "chatviewtextprocessor.h"
 
+#include <QAbstractTextDocumentLayout>
 #include <QBrush>
 #include <QChar>
 #include <QClipboard>
+#include <QDebug>
+#include <QFlag>
 #include <QFont>
-#include <QFontMetricsF>
 #include <QGuiApplication>
-#include <QList>
+#include <QList> // IWYU pragma: keep
-#include <QPainter>
+#include <QPair>
 #include <QQuickTextDocument>
 #include <QRegularExpression>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
-#include <QTextBlock>
+#include <QTextBlock> // IWYU pragma: keep
-#include <QTextCharFormat>
+#include <QTextCharFormat> // IWYU pragma: keep
 #include <QTextCursor>
 #include <QTextDocument>
 #include <QTextDocumentFragment>
-#include <QTextFrame>
+#include <QTextFrame> // IWYU pragma: keep
-#include <QTextFrameFormat>
+#include <QTextFrameFormat> // IWYU pragma: keep
 #include <QTextTableCell>
-#include <QVariant>
+#include <QtAssert>
-#include <Qt>
+#include <QtLogging>
-#include <QtGlobal>
 
 #include <algorithm>
+#include <utility>
+
 
 enum Language {
     None,

gpt4all-chat/src/chatviewtextprocessor.h

@@ -3,18 +3,15 @@
 
 #include <QColor>
 #include <QObject>
-#include <QQmlEngine>
+#include <QQmlEngine> // IWYU pragma: keep
-#include <QQuickTextDocument> // IWYU pragma: keep
+#include <QQuickTextDocument>
-#include <QRectF>
-#include <QSizeF>
 #include <QString>
 #include <QSyntaxHighlighter>
-#include <QTextObjectInterface>
+#include <QVector> // IWYU pragma: keep
-#include <QVector>
+#include <QtTypes>
+
+// IWYU pragma: no_forward_declare QQuickTextDocument
 
-class QPainter;
-class QTextDocument;
-class QTextFormat;
 
 struct CodeColors {
     Q_GADGET

gpt4all-chat/src/codeinterpreter.cpp

@@ -1,12 +1,16 @@
 #include "codeinterpreter.h"
 
+#include <QJSEngine>
 #include <QJSValue>
-#include <QStringList>
+#include <QList>
+#include <QStringList> // IWYU pragma: keep
 #include <QThread>
 #include <QVariant>
+#include <Qt>
 
 using namespace Qt::Literals::StringLiterals;
 
+
 CodeInterpreter::CodeInterpreter()
     : Tool()
     , m_error(ToolEnums::Error::NoError)

gpt4all-chat/src/codeinterpreter.h

@@ -4,11 +4,13 @@
 #include "tool.h"
 #include "toolcallparser.h"
 
-#include <QJSEngine>
 #include <QObject>
 #include <QString>
-#include <QtGlobal>
 #include <QThread>
+#include <QtAssert>
+
+class QJSEngine;
+
 
 class JavaScriptConsoleCapture : public QObject
 {

gpt4all-chat/src/database.cpp

@@ -1,19 +1,21 @@
 #include "database.h"
 
 #include "mysettings.h"
-#include "utils.h"
+#include "utils.h" // IWYU pragma: keep
 
 #include <duckx/duckx.hpp>
 #include <fmt/format.h>
+#include <usearch/index.hpp>
 #include <usearch/index_plugins.hpp>
 
-#include <QByteArrayView>
 #include <QDebug>
 #include <QDir>
 #include <QDirIterator>
 #include <QFile>
 #include <QFileSystemWatcher>
+#include <QFlags>
 #include <QIODevice>
+#include <QKeyValueIterator>
 #include <QRegularExpression>
 #include <QSqlError>
 #include <QSqlQuery>
@@ -22,8 +24,9 @@
 #include <QMap>
 #include <QUtf8StringView>
 #include <QVariant>
-#include <Qt>
 #include <QtLogging>
+#include <QtMinMax>
+#include <QtTypes>
 
 #include <algorithm>
 #include <cmath>
@@ -46,6 +49,7 @@ namespace us = unum::usearch;
 //#define DEBUG
 //#define DEBUG_EXAMPLE
 
+
 namespace {
 
 /* QFile that checks input for binary data. If seen, it fails the read and returns true
@@ -1111,9 +1115,9 @@ class DocumentReader {
 public:
     struct Metadata { QString title, author, subject, keywords; };
 
-    static std::unique_ptr<DocumentReader> fromDocument(const DocumentInfo &info);
+    static std::unique_ptr<DocumentReader> fromDocument(DocumentInfo info);
 
-    const DocumentInfo           &doc     () const { return *m_info; }
+    const DocumentInfo           &doc     () const { return m_info; }
     const Metadata               &metadata() const { return m_metadata; }
     const std::optional<QString> &word    () const { return m_word; }
     const std::optional<QString> &nextWord()       { m_word = advance(); return m_word; }
@@ -1123,8 +1127,8 @@ public:
     virtual ~DocumentReader() = default;
 
 protected:
-    explicit DocumentReader(const DocumentInfo &info)
+    explicit DocumentReader(DocumentInfo info)
-        : m_info(&info) {}
+        : m_info(std::move(info)) {}
 
     void postInit(Metadata &&metadata = {})
     {
@@ -1134,9 +1138,9 @@ protected:
 
     virtual std::optional<QString> advance() = 0;
 
-    const DocumentInfo     *m_info;
+    DocumentInfo           m_info;
-    Metadata                m_metadata;
+    Metadata               m_metadata;
-    std::optional<QString>  m_word;
+    std::optional<QString> m_word;
 };
 
 namespace {
@@ -1144,8 +1148,8 @@ namespace {
 #ifdef GPT4ALL_USE_QTPDF
 class PdfDocumentReader final : public DocumentReader {
 public:
-    explicit PdfDocumentReader(const DocumentInfo &info)
+    explicit PdfDocumentReader(DocumentInfo info)
-        : DocumentReader(info)
+        : DocumentReader(std::move(info))
     {
         QString path = info.file.canonicalFilePath();
         if (m_doc.load(path) != QPdfDocument::Error::None)
@@ -1185,8 +1189,8 @@ private:
 #else
 class PdfDocumentReader final : public DocumentReader {
 public:
-    explicit PdfDocumentReader(const DocumentInfo &info)
+    explicit PdfDocumentReader(DocumentInfo info)
-        : DocumentReader(info)
+        : DocumentReader(std::move(info))
     {
         QString path = info.file.canonicalFilePath();
         m_doc = FPDF_LoadDocument(path.toUtf8().constData(), nullptr);
@@ -1277,8 +1281,8 @@ private:
 
 class WordDocumentReader final : public DocumentReader {
 public:
-    explicit WordDocumentReader(const DocumentInfo &info)
+    explicit WordDocumentReader(DocumentInfo info)
-        : DocumentReader(info)
+        : DocumentReader(std::move(info))
         , m_doc(info.file.canonicalFilePath().toStdString())
     {
         m_doc.open();
@@ -1370,8 +1374,8 @@ protected:
 
 class TxtDocumentReader final : public DocumentReader {
 public:
-    explicit TxtDocumentReader(const DocumentInfo &info)
+    explicit TxtDocumentReader(DocumentInfo info)
-        : DocumentReader(info)
+        : DocumentReader(std::move(info))
         , m_file(info.file.canonicalFilePath())
     {
         if (!m_file.open(QIODevice::ReadOnly))
@@ -1412,13 +1416,13 @@ protected:
 
 } // namespace
 
-std::unique_ptr<DocumentReader> DocumentReader::fromDocument(const DocumentInfo &doc)
+std::unique_ptr<DocumentReader> DocumentReader::fromDocument(DocumentInfo doc)
 {
     if (doc.isPdf())
-        return std::make_unique<PdfDocumentReader>(doc);
+        return std::make_unique<PdfDocumentReader>(std::move(doc));
     if (doc.isDocx())
-        return std::make_unique<WordDocumentReader>(doc);
+        return std::make_unique<WordDocumentReader>(std::move(doc));
-    return std::make_unique<TxtDocumentReader>(doc);
+    return std::make_unique<TxtDocumentReader>(std::move(doc));
 }
 
 ChunkStreamer::ChunkStreamer(Database *database)
@@ -1426,12 +1430,12 @@ ChunkStreamer::ChunkStreamer(Database *database)
 
 ChunkStreamer::~ChunkStreamer() = default;
 
-void ChunkStreamer::setDocument(const DocumentInfo &doc, int documentId, const QString &embeddingModel)
+void ChunkStreamer::setDocument(DocumentInfo doc, int documentId, const QString &embeddingModel)
 {
     auto docKey = doc.key();
     if (!m_docKey || *m_docKey != docKey) {
         m_docKey         = docKey;
-        m_reader         = DocumentReader::fromDocument(doc);
+        m_reader         = DocumentReader::fromDocument(std::move(doc));
         m_documentId     = documentId;
         m_embeddingModel = embeddingModel;
         m_chunk.clear();
@@ -1441,7 +1445,8 @@ void ChunkStreamer::setDocument(const DocumentInfo &doc, int documentId, const Q
         if (m_database->m_documentIdCache.contains(documentId)) {
             QSqlQuery q(m_database->m_db);
             if (!m_database->removeChunksByDocumentId(q, documentId))
-                handleDocumentError("ERROR: Cannot remove chunks of document", documentId, doc.file.canonicalPath(), q.lastError());
+                handleDocumentError("ERROR: Cannot remove chunks of document",
+                                    documentId, m_reader->doc().file.canonicalPath(), q.lastError());
         }
     }
 }

gpt4all-chat/src/database.h

@@ -1,7 +1,7 @@
 #ifndef DATABASE_H
 #define DATABASE_H
 
-#include "embllm.h" // IWYU pragma: keep
+#include "embllm.h"
 
 #include <QByteArray>
 #include <QChar>
@@ -15,11 +15,11 @@
 #include <QSet>
 #include <QSqlDatabase>
 #include <QString>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 #include <QThread>
 #include <QUrl>
-#include <QVector>
+#include <QVector> // IWYU pragma: keep
-#include <QtGlobal>
+#include <QtAssert>
 
 #include <atomic>
 #include <cstddef>
@@ -28,7 +28,7 @@
 #include <memory>
 #include <optional>
 #include <utility>
-#include <vector>
+#include <vector> // IWYU pragma: keep
 
 using namespace Qt::Literals::StringLiterals;
 
@@ -39,6 +39,7 @@ class QSqlQuery;
 class QTextStream;
 class QTimer;
 
+
 /* Version 0: GPT4All v2.4.3, full-text search
  * Version 1: GPT4All v2.5.3, embeddings in hsnwlib
  * Version 2: GPT4All v3.0.0, embeddings in sqlite
@@ -171,7 +172,7 @@ public:
     explicit ChunkStreamer(Database *database);
     ~ChunkStreamer();
 
-    void setDocument(const DocumentInfo &doc, int documentId, const QString &embeddingModel);
+    void setDocument(DocumentInfo doc, int documentId, const QString &embeddingModel);
     std::optional<DocumentInfo::key_type> currentDocKey() const;
     void reset();
 

gpt4all-chat/src/download.cpp

@@ -10,32 +10,37 @@
 #include <QDebug>
 #include <QGlobalStatic>
 #include <QGuiApplication>
-#include <QIODevice>
+#include <QIODevice> // IWYU pragma: keep
 #include <QJsonArray>
 #include <QJsonDocument>
 #include <QJsonObject>
 #include <QJsonValue>
+#include <QKeyValueIterator>
 #include <QLocale>
 #include <QNetworkRequest>
-#include <QPair>
+#include <QPair> // IWYU pragma: keep
+#include <QRegularExpression>
+#include <QRegularExpressionMatch>
 #include <QSettings>
 #include <QSslConfiguration>
 #include <QSslSocket>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 #include <QTextStream>
 #include <QUrl>
 #include <QVariant>
-#include <QVector>
+#include <QVector> // IWYU pragma: keep
 #include <Qt>
+#include <QtAssert>
 #include <QtLogging>
+#include <QtMinMax>
 
-#include <algorithm>
 #include <compare>
 #include <cstddef>
 #include <utility>
 
 using namespace Qt::Literals::StringLiterals;
 
+
 class MyDownload: public Download { };
 Q_GLOBAL_STATIC(MyDownload, downloadInstance)
 Download *Download::globalInstance()

gpt4all-chat/src/download.h

@@ -13,10 +13,14 @@
 #include <QSslError>
 #include <QString>
 #include <QThread>
-#include <QtGlobal>
+#include <QtTypes>
 
+// IWYU pragma: no_forward_declare QFile
+// IWYU pragma: no_forward_declare QList
+// IWYU pragma: no_forward_declare QSslError
 class QByteArray;
 
+
 struct ReleaseInfo {
     Q_GADGET
     Q_PROPERTY(QString version MEMBER version)

gpt4all-chat/src/embllm.cpp

@@ -1,35 +1,35 @@
 #include "embllm.h"
 
-#include "modellist.h"
 #include "mysettings.h"
 
 #include <gpt4all-backend/llmodel.h>
 
 #include <QCoreApplication>
 #include <QDebug>
-#include <QFile>
 #include <QFileInfo>
 #include <QGuiApplication>
-#include <QIODevice>
 #include <QJsonArray>
 #include <QJsonDocument>
 #include <QJsonObject>
+#include <QJsonValue>
 #include <QList>
-#include <QMutexLocker>
+#include <QMutexLocker> // IWYU pragma: keep
 #include <QNetworkAccessManager>
 #include <QNetworkReply>
 #include <QNetworkRequest>
 #include <QUrl>
 #include <Qt>
-#include <QtGlobal>
+#include <QtAssert>
 #include <QtLogging>
 
 #include <exception>
+#include <string>
 #include <utility>
 #include <vector>
 
 using namespace Qt::Literals::StringLiterals;
 
+
 static const QString EMBEDDING_MODEL_NAME = u"nomic-embed-text-v1.5"_s;
 static const QString LOCAL_EMBEDDING_MODEL = u"nomic-embed-text-v1.5.f16.gguf"_s;
 
@@ -359,8 +359,11 @@ void EmbeddingLLMWorker::handleFinished()
     if (retrievedData.isValid() && retrievedData.canConvert<QVector<EmbeddingChunk>>())
         chunks = retrievedData.value<QVector<EmbeddingChunk>>();
 
-    QVariant response = reply->attribute(QNetworkRequest::HttpStatusCodeAttribute);
+    QVariant response;
-    Q_ASSERT(response.isValid());
+    if (reply->error() != QNetworkReply::NoError) {
+        response = reply->attribute(QNetworkRequest::HttpStatusCodeAttribute);
+        Q_ASSERT(response.isValid());
+    }
     bool ok;
     int code = response.toInt(&ok);
     if (!ok || code != 200) {

gpt4all-chat/src/embllm.h

@@ -5,10 +5,10 @@
 #include <QMutex>
 #include <QObject>
 #include <QString>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 #include <QThread>
 #include <QVariant>
-#include <QVector>
+#include <QVector> // IWYU pragma: keep
 
 #include <atomic>
 #include <vector>
@@ -16,6 +16,7 @@
 class LLModel;
 class QNetworkAccessManager;
 
+
 struct EmbeddingChunk {
     QString model; // TODO(jared): use to select model
     int folder_id;

gpt4all-chat/src/jinja_helpers.cpp

@@ -1,16 +1,11 @@
 #include "jinja_helpers.h"
 
-#include "utils.h"
-
-#include <fmt/format.h>
-
 #include <QString>
 #include <QUrl>
 
-#include <iterator>
-#include <map>
 #include <ranges>
-#include <vector>
+#include <string>
+#include <utility>
 
 namespace views  = std::views;
 using json = nlohmann::ordered_json;

gpt4all-chat/src/jinja_helpers.h

@@ -5,7 +5,11 @@
 
 #include <nlohmann/json.hpp>
 
-#include <QtGlobal>
+#include <QtTypes> // IWYU pragma: keep
+
+// IWYU pragma: no_forward_declare MessageItem
+// IWYU pragma: no_forward_declare PromptAttachment
+// IWYU pragma: no_forward_declare ResultInfo
 
 using json = nlohmann::ordered_json;
 

gpt4all-chat/src/jinja_replacements.cpp

@@ -2,6 +2,9 @@
 
 #include "jinja_replacements.h"
 
+#include <utility>
+
+
 // This is a list of prompt templates known to GPT4All and their associated replacements which are automatically used
 // instead when loading the chat template from GGUF. These exist for two primary reasons:
 // - HuggingFace model authors make ugly chat templates because they do not expect the end user to see them;
@@ -113,11 +116,31 @@ const std::unordered_map<std::string_view, std::string_view> CHAT_TEMPLATE_SUBST
     {%- elif message['role'] == 'system' %}
         {{- '<|system|>\n' + message['content'] + eos_token }}
     {%- elif message['role'] == 'assistant' %}
-        {{- '<|assistant|>\n'  + message['content'] + eos_token }}
+        {{- '<|assistant|>\n' + message['content'] + eos_token }}
     {%- endif %}
     {%- if loop.last and add_generation_prompt %}
         {{- '<|assistant|>' }}
     {%- endif %}
+{%- endfor %})TEMPLATE",
+    },
+    // granite-3.1-3b-a800m-instruct-Q4_0.gguf, granite-3.1-8b-instruct-Q4_0.gguf (nomic-ai/gpt4all#3471)
+    {
+        // original
+        R"TEMPLATE({%- if messages[0]['role'] == 'system' %}{%- set system_message = messages[0]['content'] %}{%- set loop_messages = messages[1:] %}{%- else %}{%- set system_message = "Knowledge Cutoff Date: April 2024. You are Granite, developed by IBM." %}{%- if tools and documents %}{%- set system_message = system_message + " You are a helpful AI assistant with access to the following tools. When a tool is required to answer the user's query, respond with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request. Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}{%- elif tools %}{%- set system_message = system_message + " You are a helpful AI assistant with access to the following tools. When a tool is required to answer the user's query, respond with <|tool_call|> followed by a JSON list of tools used. If a tool does not exist in the provided list of tools, notify the user that you do not have the ability to fulfill the request." %}{%- elif documents %}{%- set system_message = system_message + " Write the response to the user's input by strictly aligning with the facts in the provided documents. If the information needed to answer the question is not available in the documents, inform the user that the question cannot be answered based on the available data." %}{%- else %}{%- set system_message = system_message + " You are a helpful AI assistant." %}{%- endif %}{%- if controls and 'citations' in controls and documents %}{%- set system_message = system_message + ' In your response, use the symbols <co> and </co> to indicate when a fact comes from a document in the search result, e.g <co>0</co> for a fact from document 0. Afterwards, list all the citations with their corresponding documents in an ordered list.' %}{%- endif %}{%- if controls and 'hallucinations' in controls and documents %}{%- set system_message = system_message + ' Finally, after the response is written, include a numbered list of sentences from the response that are potentially hallucinated and not based in the documents.' %}{%- endif %}{%- set loop_messages = messages %}{%- endif %}{{- '<|start_of_role|>system<|end_of_role|>' + system_message + '<|end_of_text|> ' }}{%- if tools %}{{- '<|start_of_role|>tools<|end_of_role|>' }}{{- tools | tojson(indent=4) }}{{- '<|end_of_text|> ' }}{%- endif %}{%- if documents %}{{- '<|start_of_role|>documents<|end_of_role|>' }}{%- for document in documents %}{{- 'Document ' + loop.index0 | string + ' ' }}{{- document['text'] }}{%- if not loop.last %}{{- ' '}}{%- endif%}{%- endfor %}{{- '<|end_of_text|> ' }}{%- endif %}{%- for message in loop_messages %}{{- '<|start_of_role|>' + message['role'] + '<|end_of_role|>' + message['content'] + '<|end_of_text|> ' }}{%- if loop.last and add_generation_prompt %}{{- '<|start_of_role|>assistant' }}{%- if controls %}{{- ' ' + controls | tojson()}}{%- endif %}{{- '<|end_of_role|>' }}{%- endif %}{%- endfor %})TEMPLATE",
+        // replacement
+        R"TEMPLATE({%- if messages[0]['role'] == 'system' %}
+    {%- set system_message = messages[0]['content'] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set system_message = "Knowledge Cutoff Date: April 2024. You are Granite, developed by IBM. You are a helpful AI assistant." %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+{{- '<|start_of_role|>system<|end_of_role|>' + system_message + '<|end_of_text|> ' }}
+{%- for message in loop_messages %}
+    {{- '<|start_of_role|>' + message['role'] + '<|end_of_role|>' + message['content'] + '<|end_of_text|> ' }}
+    {%- if loop.last and add_generation_prompt %}
+        {{- '<|start_of_role|>assistant<|end_of_role|>' }}
+    {%- endif %}
 {%- endfor %})TEMPLATE",
     },
     // Hermes-3-Llama-3.2-3B.Q4_0.gguf, mistral-7b-openorca.gguf2.Q4_0.gguf
@@ -618,6 +641,70 @@ const std::unordered_map<std::string_view, std::string_view> CHAT_TEMPLATE_SUBST
 {%- if add_generation_prompt %}
     {{- '<|im_start|>assistant\n' }}
 {%- endif %})TEMPLATE",
+    },
+    // OLMoE-1B-7B-0125-Instruct-Q4_0.gguf (nomic-ai/gpt4all#3471)
+    {
+        // original
+        R"TEMPLATE({{ bos_token }}{% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'user' %}{{ '<|user|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
+'  + message['content'] + eos_token + '
+' }}{% else %}{{ '<|assistant|>
+'  + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
+' }}{% endif %}{% endfor %})TEMPLATE",
+        // replacement
+        R"TEMPLATE({{- bos_token }}
+{%- for message in messages %}
+    {%- if message['role'] == 'system' %}
+        {{- '<|system|>\n' + message['content'] + '\n' }}
+    {%- elif message['role'] == 'user' %}
+        {{- '<|user|>\n' + message['content'] + '\n' }}
+    {%- elif message['role'] == 'assistant' %}
+        {%- if not loop.last %}
+            {{- '<|assistant|>\n' + message['content'] + eos_token + '\n' }}
+        {%- else %}
+            {{- '<|assistant|>\n' + message['content'] + eos_token }}
+        {%- endif %}
+    {%- endif %}
+    {%- if loop.last and add_generation_prompt %}
+        {{- '<|assistant|>\n' }}
+    {%- endif %}
+{%- endfor %})TEMPLATE",
+    },
+    // OLMoE-1B-7B-0924-Instruct-Q4_0.gguf (nomic-ai/gpt4all#3471)
+    {
+        // original
+        R"TEMPLATE({{ bos_token }}{% for message in messages %}
+{% if message['role'] == 'system' %}
+{{ '<|system|>
+' + message['content'] }}
+{% elif message['role'] == 'user' %}
+{{ '<|user|>
+' + message['content'] }}
+{% elif message['role'] == 'assistant' %}
+{{ '<|assistant|>
+'  + message['content'] + eos_token }}
+{% endif %}
+{% if loop.last and add_generation_prompt %}
+{{ '<|assistant|>' }}
+{% endif %}
+{% endfor %})TEMPLATE",
+        // replacement
+        R"TEMPLATE({{- bos_token }}
+{%- for message in messages %}
+    {%- if message['role'] == 'system' %}
+        {{- '<|system|>\n' + message['content'] }}
+    {%- elif message['role'] == 'user' %}
+        {{- '<|user|>\n' + message['content'] }}
+    {%- elif message['role'] == 'assistant' %}
+        {{- '<|assistant|>\n' + message['content'] + eos_token }}
+    {%- endif %}
+    {%- if loop.last and add_generation_prompt %}
+        {{- '<|assistant|>' }}
+    {%- endif %}
+{%- endfor %})TEMPLATE",
     },
     // Phi-3.1-mini-128k-instruct-Q4_0.gguf (nomic-ai/gpt4all#3346)
     {

gpt4all-chat/src/llm.cpp

@@ -12,6 +12,9 @@
 #include <QSettings>
 #include <QUrl>
 #include <QtLogging>
+#include <QtSystemDetection>
+
+#include <string>
 
 #ifdef GPT4ALL_OFFLINE_INSTALLER
 #   include <QDesktopServices>
@@ -25,6 +28,7 @@
 
 using namespace Qt::Literals::StringLiterals;
 
+
 class MyLLM: public LLM { };
 Q_GLOBAL_STATIC(MyLLM, llmInstance)
 LLM *LLM::globalInstance()

gpt4all-chat/src/llm.h

@@ -3,7 +3,8 @@
 
 #include <QObject>
 #include <QString>
-#include <QtGlobal>
+#include <QtTypes>
+
 
 class LLM : public QObject
 {

gpt4all-chat/src/localdocs.cpp

@@ -5,10 +5,14 @@
 #include "mysettings.h"
 
 #include <QCoreApplication>
+#include <QDebug>
 #include <QGlobalStatic>
 #include <QGuiApplication>
+#include <QList>
 #include <QUrl>
 #include <Qt>
+#include <QtLogging>
+
 
 class MyLocalDocs: public LocalDocs { };
 Q_GLOBAL_STATIC(MyLocalDocs, localDocsInstance)

gpt4all-chat/src/localdocs.h

@@ -2,11 +2,14 @@
 #define LOCALDOCS_H
 
 #include "database.h"
-#include "localdocsmodel.h" // IWYU pragma: keep
+#include "localdocsmodel.h"
 
 #include <QObject>
 #include <QString>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
+
+// IWYU pragma: no_forward_declare LocalDocsModel
+
 
 class LocalDocs : public QObject
 {

gpt4all-chat/src/localdocsmodel.cpp

@@ -5,11 +5,11 @@
 
 #include <QDateTime>
 #include <QMap>
-#include <QVector>
+#include <QVector> // IWYU pragma: keep
-#include <QtGlobal>
 
 #include <utility>
 
+
 LocalDocsCollectionsModel::LocalDocsCollectionsModel(QObject *parent)
     : QSortFilterProxyModel(parent)
 {

gpt4all-chat/src/localdocsmodel.h

@@ -4,17 +4,19 @@
 #include "database.h"
 
 #include <QAbstractListModel>
-#include <QByteArray>
-#include <QHash>
 #include <QList>
-#include <QObject>
+#include <QObject> // IWYU pragma: keep
 #include <QSortFilterProxyModel>
 #include <QString>
-#include <QVariant>
 #include <Qt>
 
 #include <functional>
 
+class QByteArray;
+class QVariant;
+template <typename Key, typename T> class QHash;
+
+
 class LocalDocsCollectionsModel : public QSortFilterProxyModel
 {
     Q_OBJECT

gpt4all-chat/src/logger.cpp

@@ -2,8 +2,10 @@
 
 #include <QDateTime>
 #include <QDebug>
+#include <QFlags>
 #include <QGlobalStatic>
 #include <QIODevice>
+#include <QMutexLocker> // IWYU pragma: keep
 #include <QStandardPaths>
 
 #include <cstdio>
@@ -12,6 +14,7 @@
 
 using namespace Qt::Literals::StringLiterals;
 
+
 class MyLogger: public Logger { };
 Q_GLOBAL_STATIC(MyLogger, loggerInstance)
 Logger *Logger::globalInstance()
@@ -62,8 +65,11 @@ void Logger::messageHandler(QtMsgType type, const QMessageLogContext &, const QS
     }
     // Get time and date
     auto timestamp = QDateTime::currentDateTime().toString();
-    // Write message
+
     const std::string out = u"[%1] (%2): %3\n"_s.arg(typeString, timestamp, msg).toStdString();
+
+    // Write message
+    QMutexLocker locker(&logger->m_mutex);
     logger->m_file.write(out.c_str());
     logger->m_file.flush();
     std::cerr << out;

gpt4all-chat/src/logger.h

@@ -2,19 +2,24 @@
 #define LOGGER_H
 
 #include <QFile>
+#include <QMutex>
 #include <QString>
 #include <QtLogging>
 
-class Logger
-{
-    QFile m_file;
-
-    static void messageHandler(QtMsgType type, const QMessageLogContext &context, const QString &msg);
 
+class Logger {
 public:
+    explicit Logger();
+
     static Logger *globalInstance();
 
-    explicit Logger();
+private:
+    static void messageHandler(QtMsgType type, const QMessageLogContext &context, const QString &msg);
+
+private:
+    QFile  m_file;
+    QMutex m_mutex;
+
     friend class MyLogger;
 };
 

gpt4all-chat/src/macosdock.mm

@@ -2,6 +2,7 @@
 
 #include <Cocoa/Cocoa.h>
 
+
 void MacOSDock::showIcon()
 {
     [[NSApplication sharedApplication] setActivationPolicy:NSApplicationActivationPolicyRegular];

gpt4all-chat/src/main.cpp

@@ -12,18 +12,24 @@
 #include <gpt4all-backend/llmodel.h>
 #include <singleapplication.h>
 
+#include <QByteArray>
 #include <QCoreApplication>
 #include <QFont>
 #include <QFontDatabase>
+#include <QList>
 #include <QObject>
 #include <QQmlApplicationEngine>
 #include <QQmlContext>
 #include <QQuickWindow>
 #include <QSettings>
 #include <QString>
+#include <QStringList>
 #include <QUrl>
 #include <QVariant>
+#include <QWindow>
 #include <Qt>
+#include <QtAssert>
+#include <QtSystemDetection>
 
 #if G4A_CONFIG(force_d3d12)
 #   include <QSGRendererInterface>

gpt4all-chat/src/modellist.cpp

@@ -9,9 +9,11 @@
 
 #include <QChar>
 #include <QCoreApplication>
+#include <QCryptographicHash>
 #include <QDebug>
 #include <QDir>
 #include <QDirIterator>
+#include <QEvent>
 #include <QEventLoop>
 #include <QFile>
 #include <QFileInfo>
@@ -29,14 +31,15 @@
 #include <QSslConfiguration>
 #include <QSslSocket>
 #include <QStandardPaths>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 #include <QTextStream>
 #include <QTimer>
 #include <QUrl>
+#include <QtAssert>
 #include <QtLogging>
+#include <QtPreprocessorSupport>
 
 #include <algorithm>
-#include <cstddef>
 #include <iterator>
 #include <optional>
 #include <string>
@@ -499,10 +502,11 @@ bool GPT4AllDownloadableModels::filterAcceptsRow(int sourceRow,
     bool hasDescription = !description.isEmpty();
     bool isClone = sourceModel()->data(index, ModelList::IsCloneRole).toBool();
     bool isDiscovered = sourceModel()->data(index, ModelList::IsDiscoveredRole).toBool();
+    bool isOnline = sourceModel()->data(index, ModelList::OnlineRole).toBool();
     bool satisfiesKeyword = m_keywords.isEmpty();
     for (const QString &k : m_keywords)
         satisfiesKeyword = description.contains(k) ? true : satisfiesKeyword;
-    return !isDiscovered && hasDescription && !isClone && satisfiesKeyword;
+    return !isOnline && !isDiscovered && hasDescription && !isClone && satisfiesKeyword;
 }
 
 int GPT4AllDownloadableModels::count() const
@@ -2353,3 +2357,56 @@ void ModelList::handleDiscoveryItemErrorOccurred(QNetworkReply::NetworkError cod
     qWarning() << u"ERROR: Discovery item failed with error code \"%1-%2\""_s
                       .arg(code).arg(reply->errorString()).toStdString();
 }
+
+QStringList ModelList::remoteModelList(const QString &apiKey, const QUrl &baseUrl)
+{
+    QStringList modelList;
+
+    // Create the request
+    QNetworkRequest request;
+    request.setUrl(baseUrl.resolved(QUrl("models")));
+    request.setHeader(QNetworkRequest::ContentTypeHeader, "application/json");
+
+    // Add the Authorization header
+    const QString bearerToken = QString("Bearer %1").arg(apiKey);
+    request.setRawHeader("Authorization", bearerToken.toUtf8());
+
+    // Make the GET request
+    QNetworkReply *reply = m_networkManager.get(request);
+
+    // We use a local event loop to wait for the request to complete
+    QEventLoop loop;
+    connect(reply, &QNetworkReply::finished, &loop, &QEventLoop::quit);
+    loop.exec();
+
+    // Check for errors
+    if (reply->error() == QNetworkReply::NoError) {
+        // Parse the JSON response
+        const QByteArray responseData = reply->readAll();
+        const QJsonDocument jsonDoc = QJsonDocument::fromJson(responseData);
+
+        if (!jsonDoc.isNull() && jsonDoc.isObject()) {
+            QJsonObject rootObj = jsonDoc.object();
+            QJsonValue dataValue = rootObj.value("data");
+
+            if (dataValue.isArray()) {
+                QJsonArray dataArray = dataValue.toArray();
+                for (const QJsonValue &val : dataArray) {
+                    if (val.isObject()) {
+                        QJsonObject obj = val.toObject();
+                        const QString modelId = obj.value("id").toString();
+                        modelList.append(modelId);
+                    }
+                }
+            }
+        }
+    } else {
+        // Handle network error (e.g. print it to qDebug)
+        qWarning() << "Error retrieving models:" << reply->errorString();
+    }
+
+    // Clean up
+    reply->deleteLater();
+
+    return modelList;
+}

gpt4all-chat/src/modellist.h

@@ -5,25 +5,29 @@
 #include <QByteArray>
 #include <QDateTime>
 #include <QHash>
-#include <QLatin1StringView>
+#include <QLatin1StringView> // IWYU pragma: keep
 #include <QList>
 #include <QMutex>
 #include <QNetworkAccessManager>
 #include <QNetworkReply>
 #include <QObject>
-#include <QPair>
+#include <QPair> // IWYU pragma: keep
-#include <QQmlEngine>
+#include <QQmlEngine> // IWYU pragma: keep
 #include <QSortFilterProxyModel>
 #include <QSslError>
 #include <QString>
 #include <QVariant>
-#include <QVector>
+#include <QVector> // IWYU pragma: keep
 #include <Qt>
-#include <QtGlobal>
+#include <QtTypes>
 
 #include <optional>
 #include <utility>
 
+// IWYU pragma: no_forward_declare QObject
+// IWYU pragma: no_forward_declare QSslError
+class QUrl;
+
 using namespace Qt::Literals::StringLiterals;
 
 
@@ -530,6 +534,8 @@ public:
 
     Q_INVOKABLE void discoverSearch(const QString &discover);
 
+    Q_INVOKABLE QStringList remoteModelList(const QString &apiKey, const QUrl &baseUrl);
+
 Q_SIGNALS:
     void countChanged();
     void installedModelsChanged();

gpt4all-chat/src/mysettings.cpp

@@ -11,22 +11,27 @@
 #include <QFileInfo>
 #include <QGlobalStatic>
 #include <QGuiApplication>
-#include <QIODevice>
+#include <QIODevice> // IWYU pragma: keep
 #include <QMap>
 #include <QMetaObject>
 #include <QStandardPaths>
 #include <QThread>
 #include <QUrl>
-#include <QVariant>
 #include <QtLogging>
+#include <QtAssert>
 
 #include <algorithm>
 #include <string>
 #include <thread>
 #include <vector>
 
+#if !(defined(Q_OS_MAC) && defined(__aarch64__))
+    #include <cstring>
+#endif
+
 using namespace Qt::Literals::StringLiterals;
 
+
 // used only for settings serialization, do not translate
 static const QStringList suggestionModeNames { "LocalDocsOnly", "On", "Off" };
 static const QStringList chatThemeNames      { "Light", "Dark", "LegacyDark" };

gpt4all-chat/src/mysettings.h

@@ -4,20 +4,24 @@
 #include "modellist.h" // IWYU pragma: keep
 
 #include <QDateTime>
-#include <QLatin1StringView>
+#include <QLatin1StringView> // IWYU pragma: keep
 #include <QList>
 #include <QModelIndex>
 #include <QObject>
 #include <QSettings>
 #include <QString>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 #include <QTranslator>
-#include <QVector>
+#include <QVariant>
 
 #include <cstdint>
 #include <memory>
 #include <optional>
 
+// IWYU pragma: no_forward_declare QModelIndex
+class QLocale;
+
+
 namespace MySettingsEnums {
     Q_NAMESPACE
 

gpt4all-chat/src/network.cpp

@@ -8,7 +8,6 @@
 #include "localdocsmodel.h"
 #include "modellist.h"
 #include "mysettings.h"
-#include "utils.h"
 
 #include <gpt4all-backend/llmodel.h>
 
@@ -29,7 +28,6 @@
 #include <QSslSocket>
 #include <QSysInfo>
 #include <Qt>
-#include <QtGlobal>
 #include <QtLogging>
 #include <QUrl>
 #include <QUuid>
@@ -49,6 +47,7 @@ using namespace Qt::Literals::StringLiterals;
 #define STR_(x) #x
 #define STR(x) STR_(x)
 
+
 static const char MIXPANEL_TOKEN[] = "ce362e568ddaee16ed243eaffb5860a2";
 
 #ifdef __clang__
@@ -242,6 +241,12 @@ void Network::handleJsonUploadFinished()
 
     m_activeUploads.removeAll(jsonReply);
 
+    if (jsonReply->error() != QNetworkReply::NoError) {
+        qWarning() << "Request to" << jsonReply->url().toString() << "failed:" << jsonReply->errorString();
+        jsonReply->deleteLater();
+        return;
+    }
+
     QVariant response = jsonReply->attribute(QNetworkRequest::HttpStatusCodeAttribute);
     Q_ASSERT(response.isValid());
     bool ok;
@@ -449,6 +454,11 @@ void Network::handleIpifyFinished()
     QNetworkReply *reply = qobject_cast<QNetworkReply *>(sender());
     if (!reply)
         return;
+    if (reply->error() != QNetworkReply::NoError) {
+        qWarning() << "Request to" << reply->url().toString() << "failed:" << reply->errorString();
+        reply->deleteLater();
+        return;
+    }
 
     QVariant response = reply->attribute(QNetworkRequest::HttpStatusCodeAttribute);
     Q_ASSERT(response.isValid());
@@ -473,6 +483,11 @@ void Network::handleMixpanelFinished()
     QNetworkReply *reply = qobject_cast<QNetworkReply *>(sender());
     if (!reply)
         return;
+    if (reply->error() != QNetworkReply::NoError) {
+        qWarning() << "Request to" << reply->url().toString() << "failed:" << reply->errorString();
+        reply->deleteLater();
+        return;
+    }
 
     QVariant response = reply->attribute(QNetworkRequest::HttpStatusCodeAttribute);
     Q_ASSERT(response.isValid());
@@ -511,6 +526,11 @@ void Network::handleHealthFinished()
     QNetworkReply *healthReply = qobject_cast<QNetworkReply *>(sender());
     if (!healthReply)
         return;
+    if (healthReply->error() != QNetworkReply::NoError) {
+        qWarning() << "Request to" << healthReply->url().toString() << "failed:" << healthReply->errorString();
+        healthReply->deleteLater();
+        return;
+    }
 
     QVariant response = healthReply->attribute(QNetworkRequest::HttpStatusCodeAttribute);
     Q_ASSERT(response.isValid());

gpt4all-chat/src/network.h

@@ -11,7 +11,14 @@
 #include <QSslError>
 #include <QString>
 #include <QVariant>
-#include <QVector>
+#include <QVariantMap> // IWYU pragma: keep
+#include <QVector> // IWYU pragma: keep
+
+// IWYU pragma: no_forward_declare QByteArray
+// IWYU pragma: no_forward_declare QNetworkReply
+// IWYU pragma: no_forward_declare QSslError
+class QUrl;
+
 
 struct KeyValue {
     QString key;

gpt4all-chat/src/server.cpp

@@ -4,9 +4,10 @@
 #include "chatmodel.h"
 #include "modellist.h"
 #include "mysettings.h"
-#include "utils.h"
+#include "utils.h" // IWYU pragma: keep
 
 #include <fmt/format.h>
+#include <gpt4all-backend/llmodel.h>
 
 #include <QByteArray>
 #include <QCborArray>
@@ -15,32 +16,38 @@
 #include <QDateTime>
 #include <QDebug>
 #include <QHostAddress>
+#include <QHttpHeaders>
 #include <QHttpServer>
+#include <QHttpServerRequest>
 #include <QHttpServerResponder>
 #include <QJsonArray>
 #include <QJsonDocument>
 #include <QJsonObject>
 #include <QJsonValue>
 #include <QLatin1StringView>
-#include <QPair>
+#include <QPair> // IWYU pragma: keep
+#include <QTcpServer>
 #include <QVariant>
 #include <Qt>
+#include <QtAssert>
 #include <QtCborCommon>
-#include <QtGlobal>
 #include <QtLogging>
+#include <QtMinMax>
+#include <QtPreprocessorSupport>
+#include <QtTypes>
 
 #include <cstdint>
+#include <exception>
 #include <iostream>
 #include <optional>
+#include <span>
 #include <stdexcept>
 #include <string>
-#include <type_traits>
+#include <string_view>
 #include <unordered_map>
 #include <utility>
-
+#include <variant>
-#if QT_VERSION >= QT_VERSION_CHECK(6, 8, 0)
+#include <vector>
-#   include <QTcpServer>
-#endif
 
 using namespace std::string_literals;
 using namespace Qt::Literals::StringLiterals;
@@ -451,23 +458,17 @@ static QJsonObject requestFromJson(const QByteArray &request)
 void Server::start()
 {
     m_server = std::make_unique<QHttpServer>(this);
-#if QT_VERSION >= QT_VERSION_CHECK(6, 8, 0)
     auto *tcpServer = new QTcpServer(m_server.get());
-#else
-    auto *tcpServer = m_server.get();
-#endif
 
     auto port = MySettings::globalInstance()->networkPort();
     if (!tcpServer->listen(QHostAddress::LocalHost, port)) {
         qWarning() << "Server ERROR: Failed to listen on port" << port;
         return;
     }
-#if QT_VERSION >= QT_VERSION_CHECK(6, 8, 0)
     if (!m_server->bind(tcpServer)) {
         qWarning() << "Server ERROR: Failed to HTTP server to socket" << port;
         return;
     }
-#endif
 
     m_server->route("/v1/models", QHttpServerRequest::Method::Get,
         [](const QHttpServerRequest &) {
@@ -607,19 +608,12 @@ void Server::start()
         }
     );
 
-#if QT_VERSION >= QT_VERSION_CHECK(6, 8, 0)
     m_server->addAfterRequestHandler(this, [](const QHttpServerRequest &req, QHttpServerResponse &resp) {
         Q_UNUSED(req);
         auto headers = resp.headers();
         headers.append("Access-Control-Allow-Origin"_L1, "*"_L1);
         resp.setHeaders(std::move(headers));
     });
-#else
-    m_server->afterRequest([](QHttpServerResponse &&resp) {
-        resp.addHeader("Access-Control-Allow-Origin", "*");
-        return std::move(resp);
-    });
-#endif
 
     connect(this, &Server::requestResetResponseState, m_chat, &Chat::resetResponseState, Qt::BlockingQueuedConnection);
 }

gpt4all-chat/src/server.h

@@ -8,7 +8,7 @@
 #include <QHttpServerResponse>
 #include <QJsonObject>
 #include <QList>
-#include <QObject>
+#include <QObject> // IWYU pragma: keep
 #include <QString>
 
 #include <memory>

gpt4all-chat/src/tool.cpp

@@ -1,5 +1,8 @@
 #include "tool.h"
 
+#include <QDataStream>
+#include <QtTypes>
+
 #include <string>
 
 using json = nlohmann::ordered_json;

gpt4all-chat/src/tool.h

@@ -9,6 +9,8 @@
 #include <QVariant>
 #include <QtGlobal>
 
+class QDataStream;
+
 using json = nlohmann::ordered_json;
 
 

gpt4all-chat/src/toolcallparser.cpp

@@ -2,8 +2,10 @@
 
 #include "tool.h"
 
+#include <QChar>
 #include <QSet>
-#include <QtGlobal>
+#include <QtAssert>
+#include <QtTypes>
 
 #include <stdexcept>
 

gpt4all-chat/src/toolcallparser.h

@@ -4,7 +4,7 @@
 #include <QByteArray>
 #include <QList>
 #include <QString>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 
 namespace ToolEnums { enum class ParseState; }
 

gpt4all-chat/src/toolmodel.cpp

@@ -6,6 +6,7 @@
 #include <QEvent>
 #include <QGlobalStatic>
 
+
 class MyToolModel: public ToolModel { };
 Q_GLOBAL_STATIC(MyToolModel, toolModelInstance)
 ToolModel *ToolModel::globalInstance()

gpt4all-chat/src/toolmodel.h

@@ -9,7 +9,8 @@
 #include <QList>
 #include <QString>
 #include <QVariant>
-#include <QtGlobal>
+#include <QtPreprocessorSupport>
+
 
 class ToolModel : public QAbstractListModel
 {

gpt4all-chat/src/utils.h

@@ -5,7 +5,7 @@
 
 #include <QByteArray>
 #include <QJsonValue>
-#include <QLatin1StringView>
+#include <QLatin1StringView> // IWYU pragma: keep
 #include <QString>
 #include <QStringView>
 #include <QUtf8StringView>
@@ -13,8 +13,9 @@
 
 #include <initializer_list>
 #include <string_view>
-#include <utility>
+#include <utility> // IWYU pragma: keep
 
+// IWYU pragma: no_forward_declare QJsonValue
 class QJsonObject;
 
 
@@ -40,4 +41,4 @@ MAKE_FORMATTER(QVariant,        value.toString().toUtf8());
 // alternative to QJsonObject's initializer_list constructor that accepts Latin-1 strings
 QJsonObject makeJsonObject(std::initializer_list<std::pair<QLatin1StringView, QJsonValue>> args);
 
-#include "utils.inl"
+#include "utils.inl" // IWYU pragma: export

gpt4all-chat/src/utils.inl

@@ -1,5 +1,6 @@
 #include <QJsonObject>
 
+
 inline QJsonObject makeJsonObject(std::initializer_list<std::pair<QLatin1StringView, QJsonValue>> args)
 {
     QJsonObject obj;

gpt4all-chat/src/xlsxtomd.cpp

@@ -7,15 +7,16 @@
 #include <xlsxformat.h>
 #include <xlsxworksheet.h>
 
+#include <QChar>
 #include <QDateTime>
 #include <QDebug>
+#include <QLatin1StringView>
 #include <QList>
 #include <QRegularExpression>
 #include <QString>
-#include <QStringList>
+#include <QStringList> // IWYU pragma: keep
 #include <QStringView>
 #include <QVariant>
-#include <QtGlobal>
 #include <QtLogging>
 
 #include <memory>

gpt4all-chat/src/xlsxtomd.h

@@ -4,6 +4,7 @@
 class QIODevice;
 class QString;
 
+
 class XLSXToMD
 {
 public: