This page answers AI Mo Kimina Prover 72B q6_k quantization queries with explicit calculations from our model requirement dataset and compatibility speed table.
Short answer: AI Mo Kimina Prover 72B typically needs around 63GB VRAM at Q6_K, and 76GB is safer for smoother usage.
Estimated between Q4 and Q8 using a weighted interpolation toward Q8 memory footprint.
Throughput data below uses available compatibility measurements/estimates and is sorted by tokens per second for this model.
Need general guidance? Review full methodology.
| GPU | VRAM | Quantization | Speed | Compatibility | Buy |
|---|---|---|---|---|---|
| AMD Instinct MI300X | 192GB | Q8 | 107 tok/s | View full compatibility | Buy options |
| NVIDIA H200 SXM 141GB | 141GB | Q8 | 97 tok/s | View full compatibility | Buy options |
| NVIDIA H100 SXM5 80GB | 80GB | Q8 | 69 tok/s | View full compatibility | Buy options |
| AMD Instinct MI250X | 128GB | Q8 | 67 tok/s | View full compatibility | Buy options |
| NVIDIA H100 PCIe 80GB | 80GB | Q8 | 44 tok/s | View full compatibility | Buy options |
| RTX 5090 | 32GB | Q8 | 42 tok/s | View full compatibility | Buy options |
| NVIDIA A100 80GB SXM4 | 80GB | Q8 | 41 tok/s | View full compatibility | Buy options |
| AMD Instinct MI210 | 64GB | Q8 | 33 tok/s | View full compatibility | Buy options |
| NVIDIA A100 40GB PCIe | 40GB | Q8 | 32 tok/s | View full compatibility | Buy options |
| RTX 4090 | 24GB | Q8 | 25 tok/s | View full compatibility | Buy options |
| NVIDIA RTX 6000 Ada | 48GB | Q8 | 25 tok/s | View full compatibility | Buy options |
| NVIDIA L40 | 48GB | Q8 | 23 tok/s | View full compatibility | Buy options |
AI Mo Kimina Prover 72B at Q6_K is estimated to require about 63GB VRAM minimum, with 76GB recommended for smoother operation.
Start with AMD Instinct MI300X, NVIDIA H200 SXM 141GB, NVIDIA H100 SXM5 80GB and review each compatibility page for full speed and fit details.
Q6_K is a balance point between memory usage and quality. If your GPU is below 63GB, consider lower-bit quantization; if you have extra VRAM, compare Q8/FP16 options for quality-sensitive workloads.