Written by Jakub Rusinowski · Last updated July 23, 2024
Yes — comfortably
Yes, comfortably — Llama 3.1 8B Instruct at Q8_0 needs about 10.4 GB of the 24 GB usable on RTX 3090 Desktop (24 GB VRAM, 64 GB RAM), leaving ~13.6 GB spare and running at ~66.1 tok/s (estimated), with room for about 65,536 tokens of context.
Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~66.1 tok/s
| Usable memory for models | 24 GB |
| Memory bandwidth | 936 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
| Quant | Memory needed | Fits 24 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 17.9 GB | ✓ Yes | 32K | ~38.8 tok/s | 16 GB |
| Q8_0 | 10.4 GB | ✓ Yes | 64K | ~66.1 tok/s | 8.5 GB |
| Q6_K | 8.4 GB | ✓ Yes | 64K | ~80.9 tok/s | 6.6 GB |
| Q5_K_M | 7.5 GB | ✓ Yes | 128K | ~90.2 tok/s | 5.7 GB |
| Q4_K_M | 6.7 GB | ✓ Yes | 128K | ~101.1 tok/s | 4.8 GB |
| Q3_K_M | 5.3 GB | ✓ Yes | 128K | ~127.1 tok/s | 3.4 GB |
| Q2_K | 4.5 GB | ✓ Yes | 128K | ~148 tok/s | 2.6 GB |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes, comfortably — Llama 3.1 8B Instruct at Q8_0 needs about 10.4 GB of the 24 GB usable on RTX 3090 Desktop (24 GB VRAM, 64 GB RAM), leaving ~13.6 GB spare and running at ~66.1 tok/s (estimated), with room for about 65,536 tokens of context.
Q8_0 — it needs about 10.4 GB of the 24 GB available, downloads as roughly 8.5 GB, and runs at an estimated 66.1 tokens/sec with up to 64K of context.
Nothing binding — the model fits with headroom and generates at a usable speed on this hardware.
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
← Can I Run It? | Llama 3.1 Family model page | Check your hardware