Can I run this LLM?
Can a NVIDIA H100 (80GB) run GLM-4 9B?
GLM · MIT · estimated for a 80.0 GB machine at 4K context.
Yes — it runs great.
Best fit: GLM-4 9B at FP16 · 18.8 GB · 96–144 tok/s on a NVIDIA H100 (80GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 3.1 GB | 248–372 tok/s | Runs great |
| Q3_K_M | 4.1 GB | 226–339 tok/s | Runs great |
| Q4_K_M | 5.7 GB | 196–294 tok/s | Runs great |
| Q5_K_M | 6.7 GB | 182–273 tok/s | Runs great |
| Q6_K | 7.8 GB | 169–253 tok/s | Runs great |
| Q8_0 | 10.0 GB | 146–220 tok/s | Runs great |
| FP16 | 18.8 GB | 96–144 tok/s | Runs great |
Open the full calculator →Everything a NVIDIA H100 (80GB) runsGLM-4 9B on other machinesWhat to buy to run it →
Ran GLM-4 9B on a NVIDIA H100 (80GB)? Add your real numbers →