• This project is actively being developed thanks to the people who support on Boosty and Patreon. By providing active support, you receive enhanced AI models.

⚠️ WARNING: TensorRT version 10 does not support the Pascal architecture (10 series graphics cards). Use only with GPUs of at least the 20 series.

How to Use

1. Download the Latest Release

   • Download the latest release from here (v2.4).

2. Unpack Aimbot

   • Extract the contents of the Aimbot.

3. First Launch and Model Export

   • Run ai.exe and wait until the standard .onnx model is exported, usually taking no more than two minutes.
   • To export another model, simply place it in .onnx format in the models folder. Then, in the AI tab, select this model, and it will be exported automatically.

4. Settings

   • After successfully exporting the model, you can configure the program.
   • All settings are available in the overlay (default key is Home).
   • A list of settings can be found in the config documentation.

5. Controls

   • Right Mouse Button: Aim at the detected target.
   • F2: Exit the program.
   • F3: Activate pause for aiming.
   • F4: Reload config.
   • Home: Show overlay.

🛠 Build the Project from Source

ℹ️ NOTE: This guide is intended for advanced users. If you encounter errors while building the modules, please report them on the Discord server.

1. Install Visual Studio 2019 Community
   Download and install from the official website.

2. Install Windows SDK
   Ensure you have Windows SDK version 10.0.26100.0 installed.

3. Install CUDA and cuDNN

   • CUDA 12.4
     Download from NVIDIA CUDA Toolkit.
   • cuDNN 9.1
     Available on the NVIDIA cuDNN archive website.

4. Set Up Project Structure
   Create a folder named modules in the directory sunone_aimbot_cpp/sunone_aimbot_cpp/modules.

5. Build OpenCV with CUDA Support

   • Download and install CMake.
   • Download OpenCV.
   • Download OpenCV Contrib.
   • Create new directories: sunone_aimbot_cpp/modules/opencv/ and sunone_aimbot_cpp/modules/opencv/build.
   • Extract opencv-4.10.0 to sunone_aimbot_cpp/modules/opencv/opencv-4.10.0 and opencv_contrib-4.10.0 to sunone_aimbot_cpp/modules/opencv/opencv_contrib-4.10.0.
   • Extract cuDNN to sunone_aimbot_cpp/modules/cudnn.
   • Open CMake and set the source code location to sunone_aimbot_cpp/modules/opencv/opencv-4.10.0.
   • Set the build directory to sunone_aimbot_cpp/modules/opencv/build.
   • Click Configure.
   • Uncheck:
     • WITH_NVCUVENC
     • WITH_NVCUVID
   • Check or configure:
     • WITH_CUDA
     • WITH_CUBLAS
     • ENABLE_FAST_MATH
     • CUDA_FAST_MATH
     • WITH_CUDNN
     • CUDNN_LIBRARY = <full path>/sunone_aimbot_cpp/modules/cudnn/lib/x64/cudnn.lib
     • CUDNN_INCLUDE_DIR = <full path>/sunone_aimbot_cpp/modules/cudnn/include
     • CUDA_ARCH_BIN = Visit the CUDA Wiki to find your Nvidia GPU architecture. For example, for RTX 3080ti, enter 8.6.
     • OPENCV_DNN_CUDA
     • OPENCV_EXTRA_MODULES_PATH = <full path>/sunone_aimbot_cpp/modules/opencv/opencv_contrib-4.10.0/modules
     • BUILD_opencv_world
   • Click Configure again and ensure that the flags for CUDA_FAST_MATH and ENABLE_FAST_MATH are not reset.
   • Click Generate to build the C++ solution.
   • Close CMake and open sunone_aimbot_cpp/modules/opencv/build/OpenCV.sln.
   • Switch the build configuration to x64 and Release.
   • Open the CMakeTargets folder in the solution.
   • Right-click on ALL_BUILD and select Build. (Building the project can take up to two hours.)
   • After building, right-click on INSTALL and select Build.
   • Verify the built files exist in the following folders:
     • sunone_aimbot_cpp/modules/opencv/build/install/include/opencv2 - Contains .hpp and .h files.
     • sunone_aimbot_cpp/modules/opencv/build/install/x64/vc16/bin - Contains .dll files.
     • sunone_aimbot_cpp/modules/opencv/build/install/x64/vc16/lib - Contains .lib files.

6. Download Required Libraries

   • Boost
   • TensorRT from Yandex or NVIDIA Developer

7. Extract Libraries
   Extract the downloaded libraries into the respective directories:

   • sunone_aimbot_cpp/sunone_aimbot_cpp/modules/boost_1_82_0
   • sunone_aimbot_cpp/sunone_aimbot_cpp/modules/TensorRT-10.6.0.26

8. Compile Boost Libraries

   • Navigate to the Boost directory:

     cd sunone_aimbot_cpp/sunone_aimbot_cpp/modules/boost_1_82_0
     

   • Run the bootstrap script (from PowerShell):

     bootstrap.bat vc142
     

   • After successful bootstrapping, build Boost:

     b2.exe --build-type=complete link=static runtime-link=static threading=multi variant=release
     

9. Verify Folder Structure
   Ensure your project hierarchy looks like this:

   sunone_aimbot_cpp/
   ├── .gitattributes
   ├── .gitignore
   ├── LICENSE
   ├── README.md
   ├── sunone_aimbot_cpp.sln
   ├── include/
   ├── models/
   ├── modules/
   │   ├── boost_1_82_0/
   │   ├── imgui-1.91.2/
   │   ├── opencv/
   │   │   ├── build/
   │   │   ├── opencv_contrib-4.10.0/
   │   │   └── opencv-4.10.0/
   │   ├── stb/
   │   ├── TensorRT-10.6.0.26/
   │   └── tools/
   ├── scr/
   ├── screenshots/
   ├── config.ini
   └── ghub_mouse.dll
   

10. Configure Project Settings

    • Open the project in Visual Studio.
    • Ensure all library paths are correctly set in Project Properties under Library Directories.
    • Go to NuGet packages and install Microsoft.Windows.CppWinRT.

11. Verify CUDA Integration

    • Right-click on the project in Visual Studio.
    • Navigate to Build Dependencies > Build Customizations.
    • Ensure that CUDA 12.4 (.targets, .props) is included.

12. Build the Project

    • Switch the build configuration to Release.
    • Build the project by selecting Build > Build Solution.

Old releases

- Stored [here](https://disk.yandex.ru/d/m0jbkiLEFvnZKg)

📋 Config Documentation

• The config documentation is available in a separate repository.

📚 References

• TensorRT Documentation
• OpenCV Documentation
• Windows SDK
• Boost
• ImGui
• CppWinRT
• Python AI AIMBOT

📄 Licenses

Boost

• License: Boost Software License 1.0

OpenCV

• License: Apache License 2.0

ImGui

• License: MIT License