Can I run this LLM?
Can a NVIDIA RTX 3090 (24GB) run GLM-4.7 Flash (MoE)?
GLM · MIT · estimated for a 24.0 GB machine at 4K context.
Yes — it runs great.
Best fit: GLM-4.7 Flash (MoE) at Q4_K_M · 18.2 GB · 167–250 tok/s on a NVIDIA RTX 3090 (24GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 9.7 GB | 209–314 tok/s | Runs great |
| Q3_K_M | 12.9 GB | 191–286 tok/s | Runs great |
| Q4_K_M | 18.2 GB | 167–250 tok/s | Runs great |
| Q5_K_M | 21.3 GB | 155–232 tok/s | Won't fit |
| Q6_K | 24.7 GB | 144–216 tok/s | Won't fit |
| Q8_0 | 31.8 GB | 125–188 tok/s | Won't fit |
| FP16 | 59.9 GB | 83–125 tok/s | Won't fit |
Open the full calculator →Everything a NVIDIA RTX 3090 (24GB) runsGLM-4.7 Flash (MoE) on other machinesWhat to buy to run it →
Ran GLM-4.7 Flash (MoE) on a NVIDIA RTX 3090 (24GB)? Add your real numbers →