AlexsJones/llmfit

Detects hardware specs (CPU, RAM, GPU/VRAM) and scores models across quality, speed, fit, and context dimensions to recommend which ones run efficiently on your machine. Supports dynamic quantization selection, MoE architectures, multi-GPU setups, and integrates with local runtimes (Ollama, llama.cpp, MLX, Docker Model Runner, LM Studio). Offers both an interactive TUI with vim-like modes and visual model comparison tools, plus CLI output for scripting.

15,685 stars and 4,091 monthly downloads. Actively maintained with 173 commits in the last 30 days.