Qwen 2.5 72B Instruct — VRAM, Speed & Local Setup

作者： Jakub Rusinowski · 最后更新： 2024年9月18日

Model library → Qwen 2.5 Family → Qwen 2.5 72B Instruct

The flagship Qwen 2.5 model. Rivals GPT-4o on many benchmarks. Requires 48GB+ VRAM or dual 3090/4090 setup. The best open-weight model for demanding coding and reasoning tasks.

Qwen 2.5 72B Instruct needs about 44 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache. On Apple Silicon that figure comes out of unified memory.

Specifications

Parameters72 Billion
Context window128,000
ArchitectureDense
ProviderAlibaba Cloud
LicenceApache 2.0
Specified atQ4_K_M
System RAM80 GB
Record updated2024-09-18

Licence

Apache-2.0 — commercial use permitted. Commercial use permitted. No usage restrictions beyond attribution.

VRAM and Speed by Quantization

Modelled on a reference NVIDIA RTX 4090 (24 GB), at 8K context. Speed figures are ESTIMATES from the memory-bandwidth roofline described on the methodology page, not benchmarks we ran — rows marked measured come from published or reader-submitted runs. VRAM here includes the KV cache, so it reads higher than the headline figure above, which does not.

QuantWeightsVRAM neededEst. speedFit on 24 GB
Q2_K23.7 GB27.2 GB~4 tok/s (est.)Offloads to system RAM (slow)
Q3_K_M30.7 GB34.2 GB~3 tok/s (est.)Offloads to system RAM (slow)
Q4_K_M43.5 GB47.0 GB~2 tok/s (est.)Offloads to system RAM (slow)
Q5_K_M51.0 GB54.5 GB~2 tok/s (est.)Offloads to system RAM (slow)
Q6_K59.0 GB62.5 GB—Won't fit
Q8_076.5 GB80.0 GB—Won't fit
F16144.0 GB147.5 GB—Won't fit

Want the memory numbers alone, at every quantization level and your own context length? Use the Qwen 2.5 72B Instruct VRAM calculator.

Buy This HardwareApple MacBook Pro M5 Pro — 64 GB VRAM · 30 W board powerDeploy in the Cloud NowNVIDIA A40 on RunPod — from $0.44/hr · rate checked 2026-08

or compare on Vast.ai

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.

Recommended GPU

The cheapest catalogued GPU that runs Qwen 2.5 72B Instruct is the Apple M5 Pro (64 GB).

联盟营销声明: 本页部分链接为联盟推广链接——如果你通过它们购买，LLM Configurator 可能会获得佣金，而你无需支付任何额外费用。作为亚马逊联盟成员（Amazon Associate），LLM Configurator 会从符合条件的购买中获得收益。
Apple MacBook Pro M5 Pro
64 GB VRAM · 30 W board power
2026年价格波动较大——请以当前商品页价格为准。
在亚马逊查看价格

How to Run Qwen 2.5 72B Instruct

Install Ollama, then run:

ollama run qwen2.5:72b

Weights on Hugging Face: Qwen/Qwen2.5-72B-Instruct.

Best for: coding, reasoning, math, enterprise.

Can I Run Qwen 2.5 72B Instruct on My GPU?

Other Qwen 2.5 Family Sizes

Qwen 2.5 72B Instruct — Frequently Asked Questions

How much VRAM does Qwen 2.5 72B Instruct need?
About 44 GB at Q4_K_M — quantized weights plus framework overhead, before any KV cache. The cache grows with context length and is added on top; the table above folds it in. Apple Silicon counts unified memory toward the same figure.
Does Qwen 2.5 72B Instruct run on an RTX 4090 (24 GB)?
No. Qwen 2.5 72B Instruct needs about 44 GB at Q4_K_M, more than a single RTX 4090's 24 GB. It needs a larger card, several GPUs, or Apple Silicon with enough unified memory — or it runs with part of the weights offloaded to system RAM, which is much slower.
How do I run Qwen 2.5 72B Instruct locally?
Install Ollama and run `ollama run qwen2.5:72b`. That pulls the weights and starts a local OpenAI-compatible endpoint; after the download nothing leaves the machine.
What other sizes does Qwen 2.5 Family come in?
Qwen 2.5 Coder 32B (20 GB), Qwen 2.5 14B Instruct (9 GB), Qwen 2.5 7B Instruct (5 GB), Qwen 2.5 72B Instruct (44 GB). Every size shares the family's training and licence; the larger ones score higher and need proportionally more memory.

← All Qwen 2.5 Family models | VRAM calculator | Check your own hardware