Will it run? / Qwen2.5-VL 72B / Mac mini M4 (16GB)

Can the Mac mini M4 (16GB) run Qwen2.5-VL 72B?

Won't fit No. Qwen2.5-VL 72B needs about 50.5 GB of memory, and the Mac mini M4 (16GB) can give a model about 10.7 GB. You'd need a machine with more memory, or a smaller model.

–estimated tokens/sec
50.5 GBmemory needed
10.7 GBavailable to the model
120 GB/smemory bandwidth

Check the Mac mini M4 (16GB) price →

How we got this number

Each generated token reads the whole model from memory once: 49 GB. The Mac mini M4 (16GB) moves 120 GB/s, so the ceiling is 2.4 tokens/sec. Real machines reach a fraction of that ceiling; we use 80%, calibrated against a measured Mac run. Send us your real numbers and we'll replace the estimate.

Faster machines for Qwen2.5-VL 72B

Mac Studio M5 Ultra (96GB)20 tok/s$5,499Check price →
Mac Studio M5 Ultra (256GB)20 tok/s$9,499Check price →
Mac Studio M5 Max (40-core GPU, 64GB)6.0 tok/s$3,799Check price →
NVIDIA DGX Spark (128GB)3.9 tok/sCheck price →

Other models on the Mac mini M4 (16GB)

starCoder 2 15BUsable, but slow5.8 tok/s