Can I run this LLM?
Can a NVIDIA RTX 3090 (24GB) run Gemma 3 12B?
Gemma · Gemma · estimated for a 24.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Gemma 3 12B at Q8_0 · 12.5 GB · 43–65 tok/s on a NVIDIA RTX 3090 (24GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q2_K | 4.8 GB | 84–125 tok/s | Runs great |
| Q3_K_M | 6.0 GB | 73–109 tok/s | Runs great |
| Q4_K_M | 7.3 GB | 64–96 tok/s | Runs great |
| Q5_K_M | 8.5 GB | 58–87 tok/s | Runs great |
| Q6_K | 9.7 GB | 53–79 tok/s | Runs great |
| Q8_0 | 12.5 GB | 43–65 tok/s | Runs great |
| FP16 | 23.5 GB | 26–38 tok/s | Won't fit |
Open the full calculator →Everything a NVIDIA RTX 3090 (24GB) runsGemma 3 12B on other machinesWhat to buy to run it →
Ran Gemma 3 12B on a NVIDIA RTX 3090 (24GB)? Add your real numbers →