Can a Apple MacBook Pro (M3 Max) run Qwen 2.5 (72B) locally?

Not comfortably — the Apple MacBook Pro (M3 Max) has 36GB VRAM and Qwen 2.5 (72B) needs more than that even in its smallest quantization. Use the cloud or a larger GPU.

What is the best Qwen 2.5 (72B) quantization for a 36GB GPU?

The Q4_K_M (4-bit Recommended) variant (~43.2GB) is the best fit on the Apple MacBook Pro (M3 Max).

GPU × Model Compatibility

Can a Apple MacBook Pro (M3 Max) run Qwen 2.5 (72B)?

Apple MacBook Pro (M3 Max) has 36GB VRAM. Here's exactly whichprecision (FP16, FP8, GGUF)of Qwen 2.5 (72B) fits and how smoothly it runs.

Too much for this GPU

The Apple MacBook Pro (M3 Max) (36GB) doesn't have enough VRAM for Qwen 2.5 (72B) even at its smallest size (43.2GB needed). Rent a cloud GPU or step up to a bigger card.

43.2GB

smallest fit on this GPU

Run it in the cloud instead

A 24GB RTX 4090 instance runs Qwen 2.5 (72B) at full quality, starting at pennies/hour.

Disclosure: Cloud provider links are referrals at no additional cost to you.

RunPod ↗Vast.ai ↗

Qwen 2.5 (72B) variants on the Apple MacBook Pro (M3 Max)

Variant	Precision	Size	Total VRAM needed	Headroom	Fits?
Q4_K_M (4-bit Recommended)REC	q4_k_m	43.2GB	43.2GB	-7.2GB	✗ OOM
Q8_0 (8-bit GGUF)	q8_0	81.2GB	81.2GB	-45.2GB	✗ OOM
FP16 (Uncompressed)	fp16	152.2GB	152.2GB	-116.2GB	✗ OOM

"Total VRAM needed" = variant size + 0.0GB estimated text-encoder/VAE overhead. Headroom = GPU VRAM (36GB) − total needed. Green ≥ 1.5GB headroom; Red ≤ −2GB (will OOM).

Apple MacBook Pro (M3 Max) at a glance

VRAM

36GB

Est. price

$3500

Best for

145.4GB Qwen 2.5 (72B) and lighter

Can a Apple MacBook Pro (M3 Max) run Qwen 2.5 (72B)?

Too much for this GPU

Run it in the cloud instead

Qwen 2.5 (72B) variants on the Apple MacBook Pro (M3 Max)

Apple MacBook Pro (M3 Max) at a glance

Other models on the Apple MacBook Pro (M3 Max)

Other GPUs for Qwen 2.5 (72B)