Text model · Qwen2.5
Qwen2.5 72B requirements
Qwen2.5 family · 72B params · released Sep 2024 · 23.2M Ollama pulls · LMArena Elo 1303. Minimum to run at Q4_K_M: Apple M4 Max (128GB).
Quantization sizes
| Quantization | Size on disk |
|---|---|
| Q2_K | 30.2 GB est |
| Q3_K_M | 35.2 GB est |
| Q4_K_M (default) | 47.42 GB |
| Q5_K_M | 51.3 GB est |
| Q6_K | 59 GB est |
| Q8_0 | 77.26 GB |
| FP16 | 144 GB est |
Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.
Run it
ollama run qwen2.5:72b llama-cli -hf bartowski/Qwen2.5-72B-Instruct-GGUF:Q4_K_M lms get bartowski/Qwen2.5-72B-Instruct-GGUF Which devices can run Qwen2.5 72B?
Apple Silicon Macs
RAM-only laptops
iPhone & iPad
Android
NVIDIA GPUs
AMD GPUs
FAQ
How much VRAM or RAM does Qwen2.5 72B need?
At Q4_K_M, Qwen2.5 72B needs about 50.2 GB (weights ~47.42 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~80.1 GB.
Can Qwen2.5 72B run on a laptop?
Qwen2.5 72B is large; you need a high-memory Mac or multi-GPU setup at Q4_K_M.
Can I use Qwen2.5 72B commercially?
Conditionally. Qwen License: free for commercial use under 100M monthly active users.
Ollama shows 47GB (rounded); bartowski HF repo gives 47.42GB Q4_K_M and 77.26GB Q8_0. Default context 128K. Requires multi-GPU or high-VRAM single GPU setup.
Sources
Memory figures are estimates. See methodology.