Can I run this LLM?
Can a NVIDIA RTX 5070 Ti (16GB) run Qwen3 32B?
Qwen · Apache 2.0 · estimated for a 16.0 GB machine at 4K context.
Yes, but it's tight.
Best fit: Qwen3 32B at Q2_K · 12.3 GB · 43–65 tok/s on a NVIDIA RTX 5070 Ti (16GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 12.3 GB | 43–65 tok/s | Tight / slow |
| Q3_K_M | 16.0 GB | 35–53 tok/s | Won't fit |
| Q4_K_M | 19.8 GB | 29–44 tok/s | Won't fit |
| Q5_K_M | 23.2 GB | 25–38 tok/s | Won't fit |
| Q6_K | 26.9 GB | 22–33 tok/s | Won't fit |
| Q8_0 | 34.8 GB | 18–26 tok/s | Won't fit |
| FP16 | 65.5 GB | 10–15 tok/s | Won't fit |
Open the full calculator →Everything a NVIDIA RTX 5070 Ti (16GB) runsQwen3 32B on other machinesWhat to buy to run it →
Ran Qwen3 32B on a NVIDIA RTX 5070 Ti (16GB)? Add your real numbers →