Can the NVIDIA RTX 3090 run Llama 3.1 70B?
No
No. Even at Q4_K_M, Llama 3.1 70B needs about 46.1 GB, but the NVIDIA RTX 3090 has 24 GB usable. You could offload some layers to system RAM, but it will be very slow.
Try other context lengths in the calculator →Run it in the cloud
Too big for the NVIDIA RTX 3090? Run Llama 3.1 70B on a rented GPU instead.
Rent a cloud GPU by the hour. No upfront cost, cancel any time.
Memory needed by quantization
At 8,192 tokens of context.
| Quant | Quality | Memory | Fits? | Speed |
|---|---|---|---|---|
| FP16 | Full quality | 144.5 GB | Too big | — |
| Q8_0 | Near-lossless | 78.3 GB | Too big | — |
| Q6_K | Excellent | 61.2 GB | Too big | — |
| Q5_K_M | Very good | 53.5 GB | Too big | — |
| Q4_K_M | Good, most popular | 46.1 GB | Too big | — |
| Q3_K_M | Noticeably degraded | 37.8 GB | Too big | — |
| Q2_K | Heavily degraded | 29.8 GB | Too big | — |
How to run Llama 3.1 70B anyway
On a rented cloud GPU with enough memory, the same command you'd use locally works:
ollama run llama3.1:70bAt Q4_K_M it needs about 46.1 GB, so pick a GPU with at least 48 GB.