Pick your GPU and instantly see which open-weight LLMs fit in VRAM, at which quantization, plus benchmark scores and estimated tokens/sec.