Can I run this LLM?
Can a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB) run GLM-5.2 (MoE)?
GLM · MIT · estimated for a 144.0 GB machine at 4K context.
Yes — it runs great.
Best fit: GLM-5.2 (MoE) at Q4_K_M · 465.8 GB · 18–27 tok/s on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q3_K_M | 342.7 GB | 27–41 tok/s | Runs great |
| Q4_K_M | 465.8 GB | 18–27 tok/s | Runs great |
| Q5_K_M | 560.8 GB | 14–22 tok/s | Won't fit |
| Q6_K | 625.9 GB | 13–19 tok/s | Won't fit |
| Q8_0 | 801.4 GB | 10–14 tok/s | Won't fit |
| FP16 | 1508.0 GB | 5–7 tok/s | Won't fit |
Open the full calculator →Everything a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB) runsGLM-5.2 (MoE) on other machinesWhat to buy to run it →
Ran GLM-5.2 (MoE) on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB)? Add your real numbers →