Can I run this LLM?
Can a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB) run MiniCPM5 2B?
MiniCPM · Apache 2.0 · estimated for a 144.0 GB machine at 4K context.
Yes — it runs great.
Best fit: MiniCPM5 2B at FP16 · 5.0 GB · 240–360 tok/s on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 1.0 GB | 328–491 tok/s | Runs great |
| Q3_K_M | 1.2 GB | 321–481 tok/s | Runs great |
| Q4_K_M | 1.6 GB | 310–466 tok/s | Runs great |
| Q5_K_M | 1.9 GB | 303–454 tok/s | Runs great |
| Q6_K | 2.1 GB | 298–447 tok/s | Runs great |
| Q8_0 | 2.7 GB | 285–427 tok/s | Runs great |
| FP16 | 5.0 GB | 240–360 tok/s | Runs great |
Open the full calculator →Everything a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB) runsMiniCPM5 2B on other machinesWhat to buy to run it →
Ran MiniCPM5 2B on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB)? Add your real numbers →