A highly optimized Gomoku (Caro/Five-in-a-Row) game against an unbeatable AI, developed in Python using Pygame. The AI utilizes the Minimax Algorithm with Alpha-Beta Pruning and an advanced heuristic evaluation function to detect complex tactical threats like double-forks.

• Advanced AI Opponent: The AI ("Terminator") thinks ahead using Minimax and Alpha-Beta pruning, aggressively blocking your moves and seizing victory.
• Super-Optimized Evaluation: Evaluates cells and lines dynamically. Recognizes critical Gomoku patterns like "Open 4," "Blocked 4," "Open 3," and devastating double-threats (Forks).
• Smooth UI: A clean and interactive graphical user interface built with Pygame. Features board highlighting, winning animations, and move tracking.
• Lightning Fast: Achieves depth 3 calculations near-instantaneously thanks to localized cell evaluation updates and dynamic move sorting heuristics (get_top_moves).
• Visual Helpers: Highlights the last played move to easily keep track of the board state.

Ensure you have Python 3.8+ installed along with Pygame.

1. Clone the repository:

   git clone https://github.com/tridpt/CaroAI.git
   cd CaroAI

2. Install dependency:

   pip install pygame

3. Run the Game:

   python main.py

• Click on an empty intersection on the board to place your piece (Red 'X').
• The AI (Blue 'O') will immediately calculate and make its move.
• Get 5 pieces in a row (horizontally, vertically, or diagonally) to win.
• At the end of the game, click the mouse or press the R key to instantly restart and play again.

Developed by Trần Đức Trí. This project was conceptualized to eliminate performance bottlenecks in traditional Gomoku AI, balancing the strategic depth of the Minimax algorithm with real-time responsiveness.