This page answers AI Mo Kimina Prover 72B fp16 queries with explicit calculations from our model requirement dataset and compatibility speed table.
Short answer: AI Mo Kimina Prover 72B typically needs around 144GB VRAM at FP16, and 173GB is safer for smoother usage.
Exact FP16 requirement from model requirement data.
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 | FP16 | 58 tok/s | View full compatibility | Buy options |
| NVIDIA H200 SXM 141GB | 141GB | FP16 | 52 tok/s | View full compatibility | Buy options |
| NVIDIA H100 SXM5 80GB | 80GB | FP16 | 38 tok/s | View full compatibility | Buy options |
| AMD Instinct MI250X | 128GB | FP16 | 36 tok/s | View full compatibility | Buy options |
| NVIDIA H100 PCIe 80GB | 80GB | FP16 | 24 tok/s | View full compatibility | Buy options |
| RTX 5090 | 32GB | FP16 | 23 tok/s | View full compatibility | Buy options |
| NVIDIA A100 80GB SXM4 | 80GB | FP16 | 22 tok/s | View full compatibility | Buy options |
| AMD Instinct MI210 | 64GB | FP16 | 18 tok/s | View full compatibility | Buy options |
| NVIDIA A100 40GB PCIe | 40GB | FP16 | 17 tok/s | View full compatibility | Buy options |
| RTX 4090 | 24GB | FP16 | 14 tok/s | View full compatibility | Buy options |
| NVIDIA RTX 6000 Ada | 48GB | FP16 | 14 tok/s | View full compatibility | Buy options |
| NVIDIA L40 | 48GB | FP16 | 13 tok/s | View full compatibility | Buy options |
AI Mo Kimina Prover 72B at FP16 is estimated to require about 144GB VRAM minimum, with 173GB 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.
FP16 is a balance point between memory usage and quality. If your GPU is below 144GB, consider lower-bit quantization; if you have extra VRAM, compare Q8/FP16 options for quality-sensitive workloads.