Can I Run Llama 3.1 Family on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Autor: Jakub Rusinowski · Ostatnia aktualizacja: 23 lipca 2024
Yes — comfortably
Yes, comfortably — Llama 3.1 8B Instruct at Q8_0 needs about 10.4 GB of the 32 GB usable on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM), leaving ~21.6 GB spare and running at ~111.5 tok/s (estimated), with room for about 131,072 tokens of context.
Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~111.5 tok/s
See what else this hardware can run →
lub porównaj w Vast.ai od $0.35/godz. (typical low · varies)
Jako partner Amazon zarabiamy na kwalifikujących się zakupach. Linki do chmurowych GPU są linkami polecającymi — możemy otrzymać prowizję bez dodatkowych kosztów dla Ciebie.
RTX 5090 Desktop (32 GB VRAM, 64 GB RAM) — what it gives a model
| Usable memory for models | 32 GB |
| Memory bandwidth | 1792 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
Llama 3.1 Family on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM): memory by quantization
| Quant | Memory needed | Fits 32 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 17.9 GB | ✓ Yes | 64K | ~68.7 tok/s | 16 GB |
| Q8_0 | 10.4 GB | ✓ Yes | 128K | ~111.5 tok/s | 8.5 GB |
| Q6_K | 8.4 GB | ✓ Yes | 128K | ~132.9 tok/s | 6.6 GB |
| Q5_K_M | 7.5 GB | ✓ Yes | 128K | ~145.7 tok/s | 5.7 GB |
| Q4_K_M | 6.7 GB | ✓ Yes | 128K | ~160.3 tok/s | 4.8 GB |
| Q3_K_M | 5.3 GB | ✓ Yes | 128K | ~193 tok/s | 3.4 GB |
| Q2_K | 4.5 GB | ✓ Yes | 128K | ~217.3 tok/s | 2.6 GB |
RTX 5090 desktop limitations
- 575 W board power — budget for a 1000 W+ PSU and the heat it puts into the room.
- Models larger than 32 GB must offload to system RAM, which costs roughly an order of magnitude in speed.
Recommended setup
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
How these numbers are calculated
- Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
- 32 GB of VRAM on the NVIDIA GeForce RTX 5090 at 1792 GB/s.
- 64 GB of system RAM available for CPU offload when a model exceeds VRAM.
- Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
- KV cache is computed from this model's published attention configuration.
FAQ
Can I run Llama 3.1 Family on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Yes, comfortably — Llama 3.1 8B Instruct at Q8_0 needs about 10.4 GB of the 32 GB usable on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM), leaving ~21.6 GB spare and running at ~111.5 tok/s (estimated), with room for about 131,072 tokens of context.
Which quantization of Llama 3.1 Family should I use on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Q8_0 — it needs about 10.4 GB of the 32 GB available, downloads as roughly 8.5 GB, and runs at an estimated 111.5 tokens/sec with up to 128K of context.
What limits Llama 3.1 Family on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)?
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Which runtime should I use?
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Other Computers
Other Models on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- Llama 3.2 Family on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- Llama 3.2 Vision on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- Llama 3.3 on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- Magistral Small on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
- MiniCPM-V on RTX 5090 Desktop (32 GB VRAM, 64 GB RAM)
Llama 3.1 Family on GPUs
- Llama 3.1 Family on NVIDIA GeForce RTX 5070
- Llama 3.1 Family on NVIDIA GeForce RTX 5060 Ti 8GB
- Llama 3.1 Family on NVIDIA GeForce RTX 5060
- Llama 3.1 Family on NVIDIA GeForce RTX 4070 Ti
What This Model Is Good At
Model & Tools
← Can I Run It? | Llama 3.1 Family model page | Check your hardware