Written by Jakub Rusinowski · Last updated December 18, 2024
Yes
Yes — Falcon 3 10B Instruct at Q8_0 needs about 13.1 GB of the 16 GB usable on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM) (~2.9 GB spare), at ~54.4 tok/s (estimated), with room for about 16,384 tokens of context.
Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~54.4 tok/s
| Usable memory for models | 16 GB |
| Memory bandwidth | 960 GB/s |
| Form factor | Desktop |
| Operating system | Windows or Linux |
| Memory upgradeable | Yes |
| Quant | Memory needed | Fits 16 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 22.7 GB | ✗ No | — | — | 20.6 GB |
| Q8_0 | 13.1 GB | ✓ Yes | 16K | ~54.4 tok/s | 10.9 GB |
| Q6_K | 10.6 GB | ✓ Yes | 32K | ~67.1 tok/s | 8.4 GB |
| Q5_K_M | 9.4 GB | ✓ Yes | 32K | ~75.1 tok/s | 7.3 GB |
| Q4_K_M | 8.4 GB | ✓ Yes | 32K | ~84.6 tok/s | 6.2 GB |
| Q3_K_M | 6.5 GB | ✓ Yes | 32K | ~107.8 tok/s | 4.4 GB |
| Q2_K | 5.5 GB | ✓ Yes | 32K | ~126.9 tok/s | 3.4 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| Falcon 3 10B Instruct | 8.4 GB | ✓ Fits | ~84.6 tok/s |
| Falcon 3 7B Instruct | 6.2 GB | ✓ Fits | ~109.2 tok/s |
| Falcon 3 3B Instruct | 3.5 GB | ✓ Fits | ~182.9 tok/s |
Ollama or llama.cpp (CUDA/ROCm) — vLLM if you need concurrent serving
Yes — Falcon 3 10B Instruct at Q8_0 needs about 13.1 GB of the 16 GB usable on RTX 5080 Desktop (16 GB VRAM, 32 GB RAM) (~2.9 GB spare), at ~54.4 tok/s (estimated), with room for about 16,384 tokens of context.
Q8_0 — it needs about 13.1 GB of the 16 GB available, downloads as roughly 10.9 GB, and runs at an estimated 54.4 tokens/sec with up to 16K 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