Can I run this LLM?
Can a NVIDIA RTX 4090 (24GB) run DeepSeek-R1-Distill 14B?
DeepSeek · Apache 2.0 · estimated for a 24.0 GB machine at 4K context.
Yes — it runs great.
Best fit: DeepSeek-R1-Distill 14B at Q8_0 · 15.7 GB · 40–60 tok/s on a NVIDIA RTX 4090 (24GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 5.8 GB | 87–130 tok/s | Runs great |
| Q3_K_M | 7.3 GB | 73–110 tok/s | Runs great |
| Q4_K_M | 9.0 GB | 63–94 tok/s | Runs great |
| Q5_K_M | 10.5 GB | 56–84 tok/s | Runs great |
| Q6_K | 12.1 GB | 50–75 tok/s | Runs great |
| Q8_0 | 15.7 GB | 40–60 tok/s | Runs great |
| FP16 | 29.6 GB | 23–34 tok/s | Won't fit |
Open the full calculator →Everything a NVIDIA RTX 4090 (24GB) runsDeepSeek-R1-Distill 14B on other machinesWhat to buy to run it →
Ran DeepSeek-R1-Distill 14B on a NVIDIA RTX 4090 (24GB)? Add your real numbers →