LlamaDeck: native Windows UI to launch and benchmark llama.cpp
LlamaDeck gives Windows users a graphical way to select llama.cpp backends, load GGUF models, and run benchmarks without command‑line hassle. It parses llama‑server help output, remembers backend settings, and records throughput while keeping prompts private. Ideal for AI enthusiasts who want a polished, self‑contained launcher that streamlines model tuning and performance comparison. Compared to raw CLI use, it offers a user‑friendly interface and automated profiling for faster iteration.
View on GitHub →ByronLeeeee/LlamaDeck