Pick 2–4 models and see exactly how much VRAM, disk space, and estimated tokens/sec each requires on your GPU, at every quantization level (Q2_K through F16).
| VRAM required | Model weights + KV cache + runtime overhead at each quant level |
| Disk size | GGUF file size on disk for the selected quantization |
| Est. tok/s | Bandwidth-roofline estimate on your selected GPU (not a benchmark) |
| Fit verdict | Great / Tight / Offload / Won't run — based on your VRAM |
| Ollama command | Copy-ready pull command for the recommended quantization |
Estimates use the formula: tok/s ≈ GPU_bandwidth × 0.70 / active_weight_GB. Numbers are modelled estimates — see "How we estimate" in the tool for the full formula.