Skip to content

Mac mini M4 Pro 24GB

Apple · M4 Pro · 24GB Unified Memory · Can run 80 models

Buy Apple Amazon
Manufacturer Apple
Unified Mem 24 GB
Chip M4 Pro
CPU Cores 14
GPU Cores 20
Neural Engine 16
Bandwidth 273 GB/s
MSRP $1,399
Released Nov 8, 2024

AI Notes

The Mac mini M4 Pro 24GB provides a significant step up in AI performance with 273 GB/s memory bandwidth. With 24GB of unified memory, it can run 13B models at full precision and 30B models with quantization. The M4 Pro chip's higher bandwidth delivers noticeably faster token generation than the base M4.

Compatible Models

Model Parameters Best Quant VRAM Used Fit Est. Speed
Qwen 3 0.6B 600M Q4_K_M 2.5 GB Runs ~109 tok/s
Qwen 3.5 0.8B 800M Q4_K_M 1.5 GB Runs ~182 tok/s
Gemma 3 1B 1B Q8_0 2 GB Runs ~137 tok/s
Llama 3.2 1B 1B Q8_0 3 GB Runs ~91 tok/s
DeepSeek R1 1.5B 1.5B Q8_0 3 GB Runs ~91 tok/s
SmolLM2 1.7B 1.7B Q8_0 2.7 GB Runs ~101 tok/s
Gemma 2 2B 2B Q8_0 4 GB Runs ~68 tok/s
Gemma 3n E2B 2B Q4_K_M 3.3 GB Runs ~83 tok/s
Gemma 4 E2B 2B Q4_K_M 4 GB Runs ~68 tok/s
Qwen 3.5 2B 2B Q4_K_M 3 GB Runs ~91 tok/s
Llama 3.2 3B 3B Q8_0 5 GB Runs ~55 tok/s
StarCoder2 3B 3B Q4_K_M 3.5 GB Runs ~78 tok/s
Phi-3 Mini 3.8B 3.8B Q8_0 5.8 GB Runs ~47 tok/s
Phi-4 Mini 3.8B 3.8B Q4_K_M 4.5 GB Runs ~61 tok/s
Gemma 3 4B 4B Q4_K_M 5 GB Runs ~55 tok/s
Gemma 3n E4B 4B Q4_K_M 4.5 GB Runs ~61 tok/s
Gemma 4 E4B 4B Q4_K_M 6 GB Runs ~46 tok/s
Qwen 3 4B 4B Q4_K_M 4.5 GB Runs ~61 tok/s
Qwen 3.5 4B 4B Q4_K_M 4.5 GB Runs ~61 tok/s
Yi 1.5 6B 6B Q4_K_M 5 GB Runs ~55 tok/s
Codestral Mamba 7B 7B Q4_K_M 6.9 GB Runs ~40 tok/s
DeepSeek R1 7B 7B Q8_0 9 GB Runs ~30 tok/s
Falcon 3 7B 7B Q4_K_M 6.8 GB Runs ~40 tok/s
InternLM 2.5 7B 7B Q4_K_M 5.5 GB Runs ~50 tok/s
Mistral 7B 7B Q8_0 9 GB Runs ~30 tok/s
OpenChat 3.5 7B 7B Q4_K_M 6.9 GB Runs ~40 tok/s
Qwen 2.5 7B 7B Q8_0 9 GB Runs ~30 tok/s
Qwen 2.5 Coder 7B 7B Q8_0 9 GB Runs ~30 tok/s
Qwen 2.5 VL 7B 7B Q4_K_M 7 GB Runs ~39 tok/s
StarCoder2 7B 7B Q4_K_M 5.5 GB Runs ~50 tok/s
WizardLM 2 7B 7B Q4_K_M 6.9 GB Runs ~40 tok/s
Aya Expanse 8B 8B Q4_K_M 6.5 GB Runs ~42 tok/s
Cogito 8B 8B Q4_K_M 7.5 GB Runs ~36 tok/s
DeepSeek R1 8B 8B Q4_K_M 7.5 GB Runs ~36 tok/s
Dolphin 3 8B 8B Q4_K_M 6 GB Runs ~46 tok/s
Granite 3.3 8B 8B Q8_0 10 GB Runs ~27 tok/s
Llama 3.1 8B 8B Q8_0 10 GB Runs ~27 tok/s
Nemotron 3 Nano 8B 8B Q4_K_M 7.5 GB Runs ~36 tok/s
Nous Hermes 2 8B 8B Q4_K_M 6 GB Runs ~46 tok/s
Qwen 3 8B 8B Q4_K_M 7.5 GB Runs ~36 tok/s
Gemma 2 9B 9B Q8_0 11 GB Runs ~25 tok/s
Qwen 3.5 9B 9B Q4_K_M 7.5 GB Runs ~36 tok/s
Yi 1.5 9B 9B Q4_K_M 6.5 GB Runs ~42 tok/s
Yi Coder 9B 9B Q4_K_M 8 GB Runs ~34 tok/s
Falcon 3 10B 10B Q4_K_M 8.5 GB Runs ~32 tok/s
Llama 3.2 Vision 11B 11B Q4_K_M 8.5 GB Runs ~32 tok/s
Gemma 3 12B 12B Q4_K_M 10.5 GB Runs ~26 tok/s
Mistral Nemo 12B 12B Q4_K_M 9.5 GB Runs ~29 tok/s
DeepSeek R1 14B 14B Q4_K_M 9.9 GB Runs ~28 tok/s
Phi-4 14B 14B Q4_K_M 9.9 GB Runs ~28 tok/s
Phi-4 Reasoning 14B 14B Q4_K_M 11 GB Runs ~25 tok/s
Qwen 2.5 14B 14B Q4_K_M 9.9 GB Runs ~28 tok/s
Qwen 2.5 Coder 14B 14B Q4_K_M 12 GB Runs ~23 tok/s
Qwen 3 14B 14B Q4_K_M 12 GB Runs ~23 tok/s
StarCoder2 15B 15B Q8_0 17 GB Runs ~16 tok/s
InternLM 2.5 20B 20B Q4_K_M 12 GB Runs ~23 tok/s
Codestral 22B 22B Q4_K_M 14.7 GB Runs ~19 tok/s
Devstral 24B 24B Q4_K_M 17 GB Runs ~16 tok/s
Magistral Small 24B 24B Q4_K_M 17 GB Runs ~16 tok/s
Mistral Small 3.1 24B 24B Q4_K_M 18 GB Runs ~15 tok/s
Gemma 4 26B 26B Q4_K_M 20 GB Runs ~14 tok/s
Gemma 2 27B 27B Q4_K_M 17.7 GB Runs ~15 tok/s
Gemma 3 27B 27B Q4_K_M 20 GB Runs ~14 tok/s
Qwen 3.5 27B 27B Q4_K_M 19 GB Runs ~14 tok/s
Nous Hermes 2 34B 34B Q4_K_M 19 GB Runs ~14 tok/s
Qwen 3.5 35B A3B 35B Q4_K_M 12 GB Runs ~23 tok/s
Qwen 3 30B-A3B (MoE) 30B Q4_K_M 22 GB Runs (tight) ~12 tok/s
Gemma 4 31B 31B Q4_K_M 22 GB Runs (tight) ~12 tok/s
Aya Expanse 32B 32B Q4_K_M 22 GB Runs (tight) ~12 tok/s
Cogito 32B 32B Q4_K_M 21.5 GB Runs (tight) ~13 tok/s
DeepSeek R1 32B 32B Q4_K_M 20.7 GB Runs (tight) ~13 tok/s
Qwen 2.5 32B 32B Q4_K_M 20.7 GB Runs (tight) ~13 tok/s
QwQ 32B 32B Q4_K_M 21.5 GB Runs (tight) ~13 tok/s
WizardCoder 33B 33B Q4_K_M 22 GB Runs (tight) ~12 tok/s
Yi 1.5 34B 34B Q4_K_M 21 GB Runs (tight) ~13 tok/s
Command R 35B 35B Q4_K_M 22.5 GB Runs (tight) ~12 tok/s
Qwen 2.5 Coder 32B 32B Q4_K_M 23 GB CPU Offload ~4 tok/s
Qwen 3 32B 32B Q4_K_M 23 GB CPU Offload ~4 tok/s
Dolphin Mixtral 8x7B 47B Q4_K_M 26 GB CPU Offload ~3 tok/s
Mixtral 8x7B 47B Q4_K_M 29.7 GB CPU Offload ~3 tok/s
25 model(s) are too large for this hardware.