Instantly calculate the maximum size of quantized language models that can fit in your available RAM, helping you optimize your models for inference.
Terminal score: 0–1000 raw, weighted across 4 dimensions. Public score: 0–10 normalized (shown in the 30-day stars chart above).
