Core Integration

Project Pal brings powerful Large Language Models (LLMs) to your desktop with zero internet required. Built on the godot-llama-cpp library, it provides a streamlined interface for local AI interaction.

• ✅ Built-in Model: Includes Qwen2.5-1.5B-Instruct (1.6GB) for immediate "out-of-the-box" testing.
• ✅ Bring Your Own Brain: Simply drop any .gguf file into the /ai_model folder to swap models in-app.
• ✅ Total Privacy: Your data never leaves your machine. No APIs, no subscriptions, no tracking.

---

Technical Details

GPU Optimized: This tool is GPU-accelerated. Developed and tested on an Nvidia RTX 3080. Performance on lower-end cards or integrated graphics may vary significantly.

Common Troubleshooting

Error: "MSVCP140.dll was not found"
This occurs if the Microsoft Visual C++ Redistributable is missing. Please install it to run the llama-cpp libraries:

Download Visual C++ Redistributable (Official Microsoft Link)

---

How to Install

1. Download and Extract the ZIP file.
2. Open the folder and run ProjectPal.exe.
3. To use custom models, place your .gguf files in the /ai_model directory before launching.