LlamaTune is a small Windows-oriented helper for testing GGUF models with the local llama.cpp binaries in the parent folder.It detects NVIDIA GPU and system memory information, discovers local models, and benchmarks several GPU-layer counts (ngl).The tool stores the best-performing configuration and can launch llama-server using that optimized setting.Built to make testing and tuning local GGUF models with llama.cpp quicker and easier.