How much VRAM does Qwen 2.5 72B need?

Qwen 2.5 72B (72B parameters) requires approximately 62.8 GB of memory with Q4_K_M quantization.

What is the best quantization for Qwen 2.5 72B?

The recommended quantization for Qwen 2.5 72B is Q4_K_M, which balances quality and memory efficiency.

Can it run?

Yes, NVIDIA A16 64GB can run Qwen 2.5 72B with a C grade (Runs with offload). Expected decode speed: 9.8 tok/s.

CUsable

Runs with offload

Using Q4_K_M in Ollama

Capabilities:

Fit status

Runs with offload

Decode

9.8 tok/s

TTFT

19683 ms

Safe context

16K

Memory

62.8 GB / 64.0 GB

Weights43.9 GB

KV Cache11.3 GB

Runtime1.2 GB

Headroom6.4 GB

Workload	Grade	Fit	Decode	TTFT	Context
Agentic Coding	D	Very compromised (needs ~5.9 GB host RAM)	9.4 tok/s	29863 ms	28K
Chat	C	Tight fit	10.7 tok/s	9910 ms	9K
Coding	C	Runs with offload	9.8 tok/s	19683 ms	16K
RAG	D	Very compromised (needs ~5.9 GB host RAM)	9.4 tok/s	37329 ms	28K
Reasoning	C	Runs with offload	10.7 tok/s	21472 ms

How Qwen 2.5 72B (72B params) fits at each quantization level on NVIDIA A16 64GB (64.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	28.1 GB	Low	D39
Q3_K_S	3	35.3 GB	Low	C41
NVFP4	4

Ollama

ollama run qwen-2.5-72b

HuggingFace

huggingface-cli download qwen-2.5-72b

Upgrade options

MacBook Pro M3 Max 128GBBudget pick

C5.5 tok/s decode

~$2,499 MSRP

Mac Studio M2 Ultra 128GBBest value

C10.6 tok/s decode

~$3,999 MSRP

NVIDIA GH200 96GBBiggest leap