Merge branch 'nomic-ai:main' into main

.circleci/continue_config.yml

@@ -27,7 +27,176 @@ jobs:
       - image: circleci/python:3.7
     steps:
       - run: echo "CircleCI pipeline triggered"
-
+  build-offline-chat-installer-macos:
+    macos:
+      xcode: 14.0.0
+    steps:
+      - checkout
+      - run:
+          name: Update Submodules
+          command: |
+            git submodule sync
+            git submodule update --init --recursive
+      - restore_cache:  # this is the new step to restore cache
+          keys:
+            - macos-qt-cache_v2
+      - run:
+          name: Installing Qt
+          command: |
+            if [ ! -d ~/Qt ]; then
+              curl -o qt-unified-macOS-x64-4.6.0-online.dmg https://gpt4all.io/ci/qt-unified-macOS-x64-4.6.0-online.dmg
+              hdiutil attach qt-unified-macOS-x64-4.6.0-online.dmg
+              /Volumes/qt-unified-macOS-x64-4.6.0-online/qt-unified-macOS-x64-4.6.0-online.app/Contents/MacOS/qt-unified-macOS-x64-4.6.0-online --no-force-installations --no-default-installations --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations --email $QT_EMAIL --password $QT_PASSWORD install qt.tools.cmake qt.tools.ifw.46 qt.tools.ninja qt.qt6.651.clang_64 qt.qt6.651.qt5compat qt.qt6.651.debug_info qt.qt6.651.addons.qtpdf qt.qt6.651.addons.qthttpserver
+              hdiutil detach /Volumes/qt-unified-macOS-x64-4.6.0-online
+            fi
+      - save_cache:  # this is the new step to save cache
+          key: macos-qt-cache_v2
+          paths:
+            - ~/Qt
+      - run:
+          name: Build
+          command: |
+            mkdir build
+            cd build
+            export PATH=$PATH:$HOME/Qt/Tools/QtInstallerFramework/4.6/bin
+            ~/Qt/Tools/CMake/CMake.app/Contents/bin/cmake \
+              -DCMAKE_GENERATOR:STRING=Ninja \
+              -DBUILD_UNIVERSAL=ON \
+              -DMACDEPLOYQT=~/Qt/6.5.1/macos/bin/macdeployqt \
+              -DGPT4ALL_OFFLINE_INSTALLER=ON \
+              -DCMAKE_BUILD_TYPE=Release \
+              -DCMAKE_PREFIX_PATH:PATH=~/Qt/6.5.1/macos/lib/cmake/Qt6 \
+              -DCMAKE_MAKE_PROGRAM:FILEPATH=~/Qt/Tools/Ninja/ninja \
+              -S ../gpt4all-chat \
+              -B .
+            ~/Qt/Tools/CMake/CMake.app/Contents/bin/cmake --build . --target all
+            ~/Qt/Tools/CMake/CMake.app/Contents/bin/cmake --build . --target install
+            ~/Qt/Tools/CMake/CMake.app/Contents/bin/cmake --build . --target package
+            mkdir upload
+            cp gpt4all-installer-* upload
+      - store_artifacts:
+          path: build/upload
+  build-offline-chat-installer-linux:
+    machine:
+      image: ubuntu-2204:2023.04.2
+    steps:
+      - checkout
+      - run:
+          name: Update Submodules
+          command: |
+            git submodule sync
+            git submodule update --init --recursive
+      - restore_cache:  # this is the new step to restore cache
+          keys:
+            - linux-qt-cache
+      - run:
+          name: Setup Linux and Dependencies
+          command: |
+            wget -qO- https://packages.lunarg.com/lunarg-signing-key-pub.asc | sudo tee /etc/apt/trusted.gpg.d/lunarg.asc
+            sudo wget -qO /etc/apt/sources.list.d/lunarg-vulkan-jammy.list http://packages.lunarg.com/vulkan/lunarg-vulkan-jammy.list
+            sudo apt update && sudo apt install -y libfontconfig1 libfreetype6 libx11-6 libx11-xcb1 libxext6 libxfixes3 libxi6 libxrender1 libxcb1 libxcb-cursor0 libxcb-glx0 libxcb-keysyms1 libxcb-image0 libxcb-shm0 libxcb-icccm4 libxcb-sync1 libxcb-xfixes0 libxcb-shape0 libxcb-randr0 libxcb-render-util0 libxcb-util1 libxcb-xinerama0 libxcb-xkb1 libxkbcommon0 libxkbcommon-x11-0 bison build-essential flex gperf python3 gcc g++ libgl1-mesa-dev libwayland-dev vulkan-sdk patchelf
+      - run:
+          name: Installing Qt
+          command: |
+            if [ ! -d ~/Qt ]; then
+              wget https://gpt4all.io/ci/qt-unified-linux-x64-4.6.0-online.run
+              chmod +x qt-unified-linux-x64-4.6.0-online.run
+              ./qt-unified-linux-x64-4.6.0-online.run --no-force-installations --no-default-installations --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations --email $QT_EMAIL --password $QT_PASSWORD install qt.tools.cmake qt.tools.ifw.46 qt.tools.ninja qt.qt6.651.gcc_64 qt.qt6.651.qt5compat qt.qt6.651.debug_info qt.qt6.651.addons.qtpdf qt.qt6.651.addons.qthttpserver qt.qt6.651.qtwaylandcompositor
+            fi
+      - save_cache:  # this is the new step to save cache
+          key: linux-qt-cache
+          paths:
+            - ~/Qt
+      - run:
+          name: Build linuxdeployqt
+          command: |
+            git clone https://github.com/nomic-ai/linuxdeployqt
+            cd linuxdeployqt && qmake && sudo make install
+      - run:
+          name: Build
+          command: |
+            set -eo pipefail
+            export CMAKE_PREFIX_PATH=~/Qt/6.5.1/gcc_64/lib/cmake
+            export PATH=$PATH:$HOME/Qt/Tools/QtInstallerFramework/4.6/bin
+            mkdir build
+            cd build
+            mkdir upload
+            ~/Qt/Tools/CMake/bin/cmake -DGPT4ALL_OFFLINE_INSTALLER=ON -DCMAKE_BUILD_TYPE=Release -S ../gpt4all-chat -B .
+            ~/Qt/Tools/CMake/bin/cmake --build . --target all
+            ~/Qt/Tools/CMake/bin/cmake --build . --target install
+            ~/Qt/Tools/CMake/bin/cmake --build . --target package
+            cp gpt4all-installer-* upload
+      - store_artifacts:
+          path: build/upload
+  build-offline-chat-installer-windows:
+    machine:
+      image: 'windows-server-2019-vs2019:2022.08.1'
+      resource_class: windows.large
+      shell: powershell.exe -ExecutionPolicy Bypass
+    steps:
+      - checkout
+      - run:
+          name: Update Submodules
+          command: |
+            git submodule sync
+            git submodule update --init --recursive
+      - restore_cache:  # this is the new step to restore cache
+          keys:
+            - windows-qt-cache
+      - run:
+          name: Installing Qt
+          command: |
+            if (-not (Test-Path C:\Qt)) {
+              Invoke-WebRequest -Uri https://gpt4all.io/ci/qt-unified-windows-x64-4.6.0-online.exe -OutFile qt-unified-windows-x64-4.6.0-online.exe
+              & .\qt-unified-windows-x64-4.6.0-online.exe --no-force-installations --no-default-installations --no-size-checking --default-answer --accept-licenses --confirm-command --accept-obligations --email ${Env:QT_EMAIL} --password ${Env:QT_PASSWORD} install qt.tools.cmake qt.tools.ifw.46 qt.tools.ninja qt.qt6.651.win64_msvc2019_64 qt.qt6.651.qt5compat qt.qt6.651.debug_info qt.qt6.651.addons.qtpdf qt.qt6.651.addons.qthttpserver
+            }
+      - save_cache:  # this is the new step to save cache
+          key: windows-qt-cache
+          paths:
+            - C:\Qt
+      - run:
+          name: Install VulkanSDK
+          command: |
+            Invoke-WebRequest -Uri https://sdk.lunarg.com/sdk/download/1.3.261.1/windows/VulkanSDK-1.3.261.1-Installer.exe -OutFile VulkanSDK-1.3.261.1-Installer.exe
+            .\VulkanSDK-1.3.261.1-Installer.exe --accept-licenses --default-answer --confirm-command install
+      - run:
+          name: Build
+          command: |
+            $Env:PATH = "${Env:PATH};C:\Program Files (x86)\Windows Kits\10\bin\x64"
+            $Env:PATH = "${Env:PATH};C:\Program Files (x86)\Windows Kits\10\bin\10.0.22000.0\x64"
+            $Env:PATH = "${Env:PATH};C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64"
+            $Env:PATH = "${Env:PATH};C:\VulkanSDK\1.3.261.1\bin"
+            $Env:PATH = "${Env:PATH};C:\Qt\Tools\QtInstallerFramework\4.6\bin"
+            $Env:LIB = "${Env:LIB};C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22000.0\ucrt\x64"
+            $Env:LIB = "${Env:LIB};C:\Program Files (x86)\Windows Kits\10\Lib\10.0.22000.0\um\x64"
+            $Env:LIB = "${Env:LIB};C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\lib\x64"
+            $Env:LIB = "${Env:LIB};C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\ATLMFC\lib\x64"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\ucrt"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\um"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\shared"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\winrt"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Windows Kits\10\include\10.0.22000.0\cppwinrt"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Auxiliary\VS\include"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\include"
+            $Env:INCLUDE = "${Env:INCLUDE};C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\ATLMFC\include"
+            mkdir build
+            cd build
+            & "C:\Qt\Tools\CMake_64\bin\cmake.exe" `
+              "-DCMAKE_GENERATOR:STRING=Ninja" `
+              "-DCMAKE_BUILD_TYPE=Release" `
+              "-DCMAKE_PREFIX_PATH:PATH=C:\Qt\6.5.1\msvc2019_64" `
+              "-DCMAKE_MAKE_PROGRAM:FILEPATH=C:\Qt\Tools\Ninja\ninja.exe" `
+              "-DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON" `
+              "-DGPT4ALL_OFFLINE_INSTALLER=ON" `
+              "-S ..\gpt4all-chat" `
+              "-B ."
+            & "C:\Qt\Tools\Ninja\ninja.exe"
+            & "C:\Qt\Tools\Ninja\ninja.exe" install
+            & "C:\Qt\Tools\Ninja\ninja.exe" package
+            mkdir upload
+            copy gpt4all-installer-win64.exe upload
+      - store_artifacts:
+          path: build/upload
   build-gpt4all-chat-linux:
     machine:
       image: ubuntu-2204:2023.04.2
@@ -163,6 +332,7 @@ jobs:
             cd build
             ~/Qt/Tools/CMake/CMake.app/Contents/bin/cmake \
               -DCMAKE_GENERATOR:STRING=Ninja \
+              -DBUILD_UNIVERSAL=ON \
               -DCMAKE_BUILD_TYPE=Release \
               -DCMAKE_PREFIX_PATH:PATH=~/Qt/6.5.1/macos/lib/cmake/Qt6 \
               -DCMAKE_MAKE_PROGRAM:FILEPATH=~/Qt/Tools/Ninja/ninja \
@@ -244,6 +414,8 @@ jobs:
           command: |
             cd gpt4all-bindings/python/
             python setup.py bdist_wheel --plat-name=manylinux1_x86_64
+      - store_artifacts:
+          path: gpt4all-bindings/python/dist
       - persist_to_workspace:
           root: gpt4all-bindings/python/dist
           paths:
@@ -274,7 +446,9 @@ jobs:
           name: Build wheel
           command: |
             cd gpt4all-bindings/python
-            python setup.py bdist_wheel --plat-name=macosx_10_9_universal2
+            python setup.py bdist_wheel --plat-name=macosx_10_15_universal2
+      - store_artifacts:
+          path: gpt4all-bindings/python/dist
       - persist_to_workspace:
           root: gpt4all-bindings/python/dist
           paths:
@@ -288,9 +462,6 @@ jobs:
       - run:
           name: Install MinGW64
           command: choco install -y mingw --force --no-progress
-      - run:
-          name: Add MinGW64 to PATH
-          command: $env:Path += ";C:\ProgramData\chocolatey\lib\mingw\tools\install\mingw64\bin"
       - run:
           name: Install VulkanSDK
           command: |
@@ -311,8 +482,9 @@ jobs:
             cd gpt4all-backend
             mkdir build
             cd build
+            $env:Path += ";C:\ProgramData\mingw64\mingw64\bin"
             $env:Path += ";C:\VulkanSDK\1.3.261.1\bin"
-            cmake -G "MinGW Makefiles" .. -DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON
+            cmake -G "MinGW Makefiles" .. -DKOMPUTE_OPT_DISABLE_VULKAN_VERSION_CHECK=ON -DKOMPUTE_OPT_USE_BUILT_IN_VULKAN_HEADER=OFF
             cmake --build . --parallel
       - run:
           name: Build wheel
@@ -323,9 +495,11 @@ jobs:
             cd gpt4all
             mkdir llmodel_DO_NOT_MODIFY
             mkdir llmodel_DO_NOT_MODIFY/build/
-            cp 'C:\ProgramData\chocolatey\lib\mingw\tools\install\mingw64\bin\*dll' 'llmodel_DO_NOT_MODIFY/build/'
+            cp 'C:\ProgramData\mingw64\mingw64\bin\*dll' 'llmodel_DO_NOT_MODIFY/build/'
             cd ..
             python setup.py bdist_wheel --plat-name=win_amd64
+      - store_artifacts:
+          path: gpt4all-bindings/python/dist
       - persist_to_workspace:
           root: gpt4all-bindings/python/dist
           paths:
@@ -442,7 +616,7 @@ jobs:
       - run:
           name: Build Libraries
           command: |
-            $MinGWBin = "C:\ProgramData\chocolatey\lib\mingw\tools\install\mingw64\bin"
+            $MinGWBin = "C:\ProgramData\mingw64\mingw64\bin"
             $Env:Path += ";$MinGwBin"
             $Env:Path += ";C:\Program Files\CMake\bin"
             $Env:Path += ";C:\VulkanSDK\1.3.261.1\bin"
@@ -682,6 +856,7 @@ jobs:
       - node/install-packages:
           app-dir: gpt4all-bindings/typescript
           pkg-manager: yarn
+          override-ci-command: yarn install
       - run:
           command: | 
             cd gpt4all-bindings/typescript
@@ -711,6 +886,7 @@ jobs:
       - node/install-packages:
           app-dir: gpt4all-bindings/typescript
           pkg-manager: yarn
+          override-ci-command: yarn install
       - run:
           command: | 
             cd gpt4all-bindings/typescript
@@ -820,14 +996,28 @@ jobs:
           command: |
             cd gpt4all-bindings/typescript
             npm set //registry.npmjs.org/:_authToken=$NPM_TOKEN
-            npm publish --access public --tag alpha
+            npm publish
 
 workflows:
   version: 2
   default:
     when: << pipeline.parameters.run-default-workflow >>
     jobs:
       - default-job
+  build-chat-offline-installers:
+    when: << pipeline.parameters.run-chat-workflow >>
+    jobs:
+      - hold:
+          type: approval
+      - build-offline-chat-installer-macos:
+          requires:
+            - hold
+      - build-offline-chat-installer-windows:
+          requires:
+            - hold
+      - build-offline-chat-installer-linux:
+          requires:
+            - hold
   build-and-test-gpt4all-chat:
     when: << pipeline.parameters.run-chat-workflow >>
     jobs:

.codespellrc

@@ -1,3 +1,3 @@
 [codespell]
-ignore-words-list = blong, belong, afterall, som
+ignore-words-list = blong, afterall, som, assistent, crasher
 skip = .git,*.pdf,*.svg,*.lock

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -27,21 +27,6 @@ body:
         - label: "The official example notebooks/scripts"
         - label: "My own modified scripts"
 
-  - type: checkboxes
-    id: related-components
-    attributes:
-      label: Related Components
-      description: "Select the components related to the issue (if applicable):"
-      options:
-        - label: "backend"
-        - label: "bindings"
-        - label: "python-bindings"
-        - label: "chat-ui"
-        - label: "models"
-        - label: "circleci"
-        - label: "docker"
-        - label: "api"
-
   - type: textarea
     id: reproduction
     validations:
@@ -67,4 +52,4 @@ body:
       required: true
     attributes:
       label: Expected behavior
-      description: "A clear and concise description of what you would expect to happen."
+      description: "A clear and concise description of what you would expect to happen."

.gitmodules

@@ -1,9 +1,4 @@
-[submodule "llama.cpp-230519"]
-	path = gpt4all-backend/llama.cpp-230519
-	url = https://github.com/ggerganov/llama.cpp.git
-[submodule "llama.cpp-230511"]
-	path = gpt4all-backend/llama.cpp-230511
-	url = https://github.com/nomic-ai/llama.cpp
 [submodule "llama.cpp-mainline"]
 	path = gpt4all-backend/llama.cpp-mainline
 	url = https://github.com/nomic-ai/llama.cpp.git
+	branch = gguf

README.md

@@ -1,9 +1,9 @@
 <h1 align="center">GPT4All</h1>
 
-<p align="center">Open-source assistant-style large language models that run locally on your CPU</p>
+<p align="center">Open-source large language models that run locally on your CPU and nearly any GPU</p>
 
 <p align="center">
-<a href="https://gpt4all.io">GPT4All Website</a>
+<a href="https://gpt4all.io">GPT4All Website and Models</a>
 </p>
 
 <p align="center">
@@ -30,13 +30,24 @@ Run on an M1 macOS Device (not sped up!)
 </p>
 
 ## GPT4All: An ecosystem of open-source on-edge large language models.
-GPT4All is an ecosystem to train and deploy **powerful** and **customized** large language models that run locally on consumer grade CPUs. Note that your CPU needs to support [AVX or AVX2 instructions](https://en.wikipedia.org/wiki/Advanced_Vector_Extensions).
+
+> [!IMPORTANT]
+> GPT4All v2.5.0 and newer only supports models in GGUF format (.gguf). Models used with a previous version of GPT4All (.bin extension) will no longer work.
+
+GPT4All is an ecosystem to run **powerful** and **customized** large language models that work locally on consumer grade CPUs and any GPU. Note that your CPU needs to support [AVX or AVX2 instructions](https://en.wikipedia.org/wiki/Advanced_Vector_Extensions).
 
 Learn more in the [documentation](https://docs.gpt4all.io).
 
-The goal is simple - be the best instruction tuned assistant-style language model that any person or enterprise can freely use, distribute and build on.
+A GPT4All model is a 3GB - 8GB file that you can download and plug into the GPT4All open-source ecosystem software. **Nomic AI** supports and maintains this software ecosystem to enforce quality and security alongside spearheading the effort to allow any person or enterprise to easily train and deploy their own on-edge large language models.
 
-A GPT4All model is a 3GB - 8GB file that you can download and plug into the GPT4All open-source ecosystem software. **Nomic AI** supports and maintains this software ecosystem to enforce quality and security alongside spearheading the effort to allow any person or enterprise to easily train and deploy their own on-edge large language models. 
+### What's New ([Issue Tracker](https://github.com/orgs/nomic-ai/projects/2))
+- **October 19th, 2023**: GGUF Support Launches with Support for:
+    - Mistral 7b base model, an updated model gallery on [gpt4all.io](https://gpt4all.io), several new local code models including Rift Coder v1.5
+    - [Nomic Vulkan](https://blog.nomic.ai/posts/gpt4all-gpu-inference-with-vulkan) support for Q4_0, Q6 quantizations in GGUF.
+    - Offline build support for running old versions of the GPT4All Local LLM Chat Client.
+- **September 18th, 2023**: [Nomic Vulkan](https://blog.nomic.ai/posts/gpt4all-gpu-inference-with-vulkan) launches supporting local LLM inference on AMD, Intel, Samsung, Qualcomm and NVIDIA GPUs.
+- **August 15th, 2023**: GPT4All API launches allowing inference of local LLMs from docker containers.
+- **July 2023**: Stable support for LocalDocs, a GPT4All Plugin that allows you to privately and locally chat with your data.
 
 
 ### Chat Client

gpt4all-api/gpt4all_api/app/api_v1/routes/engines.py

@@ -26,7 +26,7 @@ class EngineResponse(BaseModel):
 async def list_engines():
     '''
     List all available GPT4All models from
-    https://raw.githubusercontent.com/nomic-ai/gpt4all/main/gpt4all-chat/metadata/models.json
+    https://raw.githubusercontent.com/nomic-ai/gpt4all/main/gpt4all-chat/metadata/models2.json
     '''
     raise NotImplementedError()
     return ListEnginesResponse(data=[])