We focus on models like Llama 3, DeepSeek, and Mistral to support the decentralized AI movement. We prioritize VRAM capacity over gaming benchmark scores.