Qwen on Apple M4

Can I run Qwen3 235B-A22B on an Mac Mini M4 Pro 48GB?

Not at Q4_K_M — it needs 134.8 GB against 36.0 GB available. You would need 4 of these cards.

Won't fit

134.8 GB of 36.0 GB · 374%
0140 GB
Weights 132.8 GB
KV cache 1.5 GB
Runtime overhead 0.5 GB
Over the limit 98.8 GB

Short by 98.8 GB. You can run it with 24 of 94 layers on the Mac Mini M4 Pro 48GB and the rest in system RAM, at roughly 3.70 tok/s — usable for batch work, painful for chat. A smaller quantisation or a shorter context is usually the better trade.

Generation3.70tok/s
Prompt processing161tok/s
Max context0tokens
KV per 1K tokens0GB

Every quantisation of Qwen3 235B-A22B on a Mac Mini M4 Pro 48GB

Highlighted row is the highest quality that still fits at 8K context.

QuantisationbpwWeights TotalVerdictMax ctxtok/s
FP16 / BF16 16.00 437.9 GB 439.9 GB Won't fit 1.03
INT8 / W8A8 8.50 232.3 GB 234.3 GB Won't fit 1.99
Q8_0 (GGUF) 8.50 232.3 GB 234.3 GB Won't fit 1.99
FP8 (E4M3) 8.00 218.9 GB 220.9 GB Won't fit 2.12
Q6_K 6.56 179.5 GB 181.5 GB Won't fit 2.62
Q5_K_M 5.67 155.2 GB 157.2 GB Won't fit 3.09
Q5_K_S 5.52 151.1 GB 153.1 GB Won't fit 3.19
Q4_K_M 4.85 132.8 GB 134.8 GB Won't fit 3.70
Q4_K_S 4.58 125.5 GB 127.5 GB Won't fit 3.93
Q4_0 4.55 124.7 GB 126.7 GB Won't fit 3.95
AWQ 4-bit 4.25 118.0 GB 120.0 GB Won't fit 4.23
GPTQ 4-bit 4.25 118.0 GB 120.0 GB Won't fit 4.23
MXFP4 4.25 118.0 GB 120.0 GB Won't fit 4.23
IQ4_XS 4.25 116.5 GB 118.5 GB Won't fit 4.28
Q3_K_M 3.91 107.2 GB 109.2 GB Won't fit 4.71
IQ3_M 3.70 101.5 GB 103.5 GB Won't fit 5.04
IQ3_XXS 3.06 84.1 GB 86.1 GB Won't fit 6.38
Q2_K 2.63 72.4 GB 74.4 GB Won't fit 7.74
IQ2_XXS 2.06 56.9 GB 58.9 GB Won't fit 10.9
IQ1_M 1.75 48.4 GB 50.4 GB Won't fit 14.1

Also worth checking