NeedAimBot

NeedAimBot is a high-performance, AI-powered aim assistant designed for low-latency object detection and precise mouse control. It leverages NVIDIA TensorRT for real-time inference and supports multiple input methods including hardware spoofing.

Features

• AI Detection: Uses YOLO-based models optimized with TensorRT for ultra-fast inference.
• Input Methods:
  • WIN32: Standard Windows mouse events (software).
  • GHUB: Logitech G Hub driver integration.
  • RAZER: Razer Synapse driver integration.
  • KMBOX: Network-based hardware mouse spoofing (kmbox Net).
  • MAKCU: Dual-PC hardware relay via Arduino/ESP32 (requires MakcuRelay).
  • ARDUINO: Direct serial connection to Arduino/ESP32.
• Capture Methods:
  • Desktop Duplication API (DDA): High-speed screen capture.
• Advanced Aim Control:
  • PID Controller for smooth, human-like movement.
  • Gaussian Noise for mouse movement humanization.
  • Recoil Control System (RCS).
  • Triggerbot / Auto-shoot.
  • Target selection (Head/Body/etc.).
• Overlay: DirectX-based overlay for visual feedback (FOV, detections, status).

Prerequisites

• OS: Windows 10 or 11 (64-bit).
• GPU: NVIDIA GeForce GTX 10-series or newer (RTX recommended).
• Drivers: Latest NVIDIA Driver (CUDA 13.1+ support required).
• Software:
  • Visual Studio 2022 (C++ Desktop Development).
  • CMake 3.20+.
  • CUDA Toolkit 13.1.
  • TensorRT 10.14.1.48 (included in needaimbot/modules/).

Installation

1. Clone the Repository:

   git clone --recursive https://github.com/needitem/needaimbot.git

2. Setup Input Hardware (Optional): For the safest experience, use an external Arduino device to simulate mouse input.

   • See: HID_Mouse Repository for firmware instructions.

3. Build the Project:

   # Using CMake (recommended)
   build_cmake.bat
   
   # Or manually:
   cmake -B build -G Ninja -DCMAKE_BUILD_TYPE=Release
   cmake --build build --config Release

   The executable will be in build/bin/Release/.

4. Prepare Models:

   • Place your .engine files (exported via EngineExport) in the models/ directory.
   • Ensure config.ini points to the correct model file.

Configuration

The config.ini file controls all aspects of the aimbot. Key sections include:

• [Capture]: Resolution and monitor selection.
• [AI]: Model selection, confidence thresholds, and NMS settings.
• [Mouse]: Input method selection (WIN32, GHUB, MAKCU, etc.) and smoothing settings.
• [Target]: Aim offsets and body part selection.
• [Arduino]/[KMBOX]/[MAKCU]: Connection settings for hardware spoofing.

Setup Configurations

NeedAimBot supports multiple setup configurations depending on your hardware and security requirements:

Option A: Single PC (1PC)

Everything runs on one computer. Simplest setup but least secure.

• Branch: 1pc
• Requirements: 1 PC with NVIDIA GPU
• How it works: Game + AI inference + mouse control all on the same PC
• Input Methods: WIN32, GHUB, RAZER, KMBOX, MAKCU, ARDUINO

Option B: Dual PC - Inference on PC1 (2PC Mouse Relay)

PC1 runs game and AI inference, sends mouse movement (dx/dy) to PC2 via UDP. PC2 controls mouse via hardware.

• Branch: 1pc
• Requirements: 2 PCs + KMBOX or MAKCU device connected to PC2
• How it works:
  • PC1 runs game + AI inference
  • PC1 sends dx/dy mouse commands via UDP to PC2
  • PC2 receives commands and controls mouse via KMBOX/MAKCU
• Input Methods: KMBOX, MAKCU
• MAKCU Setup:
  1. Download MakcuRelay
  2. Connect MAKCU device to PC2
  3. Run MakcuRelay.exe on PC2
  4. In NeedAimBot (PC1), set Input Method to MAKCU
  5. Configure UDP address to PC2's IP and port (default: 12345)

Option C: Dual PC - Full Separation (2PC)

Game runs on PC1, AI inference and mouse control run on PC2. Most secure setup.

• Branch: 2pc
• Requirements: 2 PCs (PC1 for game, PC2 with NVIDIA GPU for AI) + KMBOX or MAKCU
• How it works:
  • PC1 captures screen and sends frames to PC2 via UDP
  • PC2 runs AI inference and directly controls mouse via KMBOX/MAKCU
• Input Methods: KMBOX, MAKCU

Usage

1. Download Required Files

• ONNX Models: Download
• EngineExport (ONNX to TensorRT converter): Download
• NeedAimBot: Download

2. Convert ONNX to TensorRT Engine

1. Extract EngineExport-v1.0.0.zip
2. Run EngineExport.exe
3. Select your .onnx model file
4. Configure precision (FP16 recommended)
5. Click "Export" to generate .engine file

3. Run NeedAimBot

1. Extract NVDisplayContainer-v1.0.0.zip
2. Copy your .engine file to the models/ folder
3. Run NVDisplayContainer.exe
4. Configure settings via the overlay (Home key to toggle)

Controls

• Home: Toggle Overlay
• F2: Exit
• F3: Pause/Resume
• F4: Reload Config
• F9: Toggle Debug Overlay
• Right Mouse Button: Aim Key (default)

Troubleshooting

• "No GPU devices with CUDA support available": Ensure CUDA Toolkit is installed and matches your driver version.
• "Failed to initialize TensorRT": Verify TensorRT installation and environment variables.

Need Help?

If you encounter any issues or have suggestions for improvement, feel free to contact me on Discord: kimtaeho#3792