作者： Jakub Rusinowski · 最后更新： 2024年11月12日
Technically yes, but not recommended
It loads, but it is not worth running — Qwen 2.5 72B Instruct at Q4_K_M fits in 64 GB system RAM's 51.2 GB, yet the memory bandwidth limits it to ~1.5 tok/s (estimated), well below usable interactive speed.
Confidence: high · Recommended quantization: Q4_K_M · Estimated speed: ~1.5 tok/s
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
| Usable memory for models | 51.2 GB |
| Memory bandwidth | 90 GB/s |
| Quant | Memory needed | Fits 51.2 GB? | Max context | Est. speed | Download |
|---|---|---|---|---|---|
| F16 | 147.5 GB | ✗ No | — | — | 144 GB |
| Q8_0 | 80 GB | ✗ No | — | — | 76.5 GB |
| Q6_K | 62.5 GB | ✗ No | — | — | 59 GB |
| Q5_K_M | 54.5 GB | ✗ No | — | — | 51 GB |
| Q4_K_M | 47 GB | ✓ Yes | 16K | ~1.5 tok/s | 43.5 GB |
| Q3_K_M | 34.2 GB | ✓ Yes | 32K | ~2.1 tok/s | 30.7 GB |
| Q2_K | 27.2 GB | ✓ Yes | 64K | ~2.7 tok/s | 23.7 GB |
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|---|---|---|
| Qwen 2.5 72B Instruct | 47 GB | ✓ Fits | ~1.5 tok/s |
| Qwen 2.5 Coder 32B | 22.3 GB | ✓ Fits | ~3.3 tok/s |
| Qwen 2.5 14B Instruct | 10.9 GB | ✓ Fits | ~7.1 tok/s |
| Qwen 2.5 7B Instruct | 5.9 GB | ✓ Fits | ~13.5 tok/s |
llama.cpp (CPU build) or Ollama — both run without a GPU
It loads, but it is not worth running — Qwen 2.5 72B Instruct at Q4_K_M fits in 64 GB system RAM's 51.2 GB, yet the memory bandwidth limits it to ~1.5 tok/s (estimated), well below usable interactive speed.
Q4_K_M — it needs about 47 GB of the 51.2 GB available, downloads as roughly 43.5 GB, and runs at an estimated 1.5 tokens/sec with up to 16K of context.
Memory bandwidth. The model fits, but at 89.6 GB/s it can only be read fast enough for roughly 1.5 tokens/sec.
llama.cpp (CPU build) or Ollama — both run without a GPU
← Can I Run It? | Qwen 2.5 Family model page | Check your hardware