MacBook Pro M2 Pro 16GB

Apple · M2 Pro · 16GB Unified Memory · Can run 36 models

Buy Apple
Manufacturer Apple
Unified Mem 16 GB
Chip M2 Pro
CPU Cores 12
GPU Cores 19
Neural Engine 16
Bandwidth 200 GB/s
MSRP $1,999
Released Jan 17, 2023

AI Notes

The MacBook Pro M2 Pro 16GB delivers double the memory bandwidth of the base M2 at 200 GB/s, resulting in significantly faster inference. It runs 7B models at high speed and handles 13B models comfortably. The Pro chip's active cooling sustains performance during extended inference sessions.

Compatible Models

Model Parameters Best Quant VRAM Used Fit Est. Speed
Gemma 3 1B 1B Q8_0 2 GB Runs ~100 tok/s
Llama 3.2 1B 1B Q8_0 3 GB Runs ~67 tok/s
DeepSeek R1 1.5B 1.5B Q8_0 3 GB Runs ~67 tok/s
Gemma 2 2B 2B Q8_0 4 GB Runs ~50 tok/s
Llama 3.2 3B 3B Q8_0 5 GB Runs ~40 tok/s
Phi-3 Mini 3.8B 3.8B Q8_0 5.8 GB Runs ~34 tok/s
Phi-4 Mini 3.8B 3.8B Q4_K_M 4.5 GB Runs ~44 tok/s
Gemma 3 4B 4B Q4_K_M 5 GB Runs ~40 tok/s
Qwen 3 4B 4B Q4_K_M 4.5 GB Runs ~44 tok/s
DeepSeek R1 7B 7B Q8_0 9 GB Runs ~22 tok/s
Mistral 7B 7B Q8_0 9 GB Runs ~22 tok/s
Qwen 2.5 7B 7B Q8_0 9 GB Runs ~22 tok/s
Qwen 2.5 Coder 7B 7B Q8_0 9 GB Runs ~22 tok/s
DeepSeek R1 8B 8B Q4_K_M 7.5 GB Runs ~27 tok/s
Llama 3.1 8B 8B Q8_0 10 GB Runs ~20 tok/s
Qwen 3 8B 8B Q4_K_M 7.5 GB Runs ~27 tok/s
Gemma 2 9B 9B Q8_0 11 GB Runs ~18 tok/s
Gemma 3 12B 12B Q4_K_M 10.5 GB Runs ~19 tok/s
Mistral Nemo 12B 12B Q4_K_M 9.5 GB Runs ~21 tok/s
DeepSeek R1 14B 14B Q4_K_M 9.9 GB Runs ~20 tok/s
Phi-4 14B 14B Q4_K_M 9.9 GB Runs ~20 tok/s
Qwen 2.5 14B 14B Q4_K_M 9.9 GB Runs ~20 tok/s
Qwen 2.5 Coder 14B 14B Q4_K_M 12 GB Runs ~17 tok/s
Qwen 3 14B 14B Q4_K_M 12 GB Runs ~17 tok/s
Codestral 22B 22B Q4_K_M 14.7 GB Runs (tight) ~14 tok/s
StarCoder2 15B 15B Q8_0 17 GB CPU Offload ~12 tok/s
Devstral 24B 24B Q4_K_M 17 GB CPU Offload ~12 tok/s
Mistral Small 3.1 24B 24B Q4_K_M 18 GB CPU Offload ~11 tok/s
Gemma 2 27B 27B Q4_K_M 17.7 GB CPU Offload ~11 tok/s
Gemma 3 27B 27B Q4_K_M 20 GB CPU Offload ~10 tok/s
Qwen 3 30B-A3B (MoE) 30B Q4_K_M 22 GB CPU Offload ~9 tok/s
DeepSeek R1 32B 32B Q4_K_M 20.7 GB CPU Offload ~10 tok/s
Qwen 2.5 32B 32B Q4_K_M 20.7 GB CPU Offload ~10 tok/s
Qwen 2.5 Coder 32B 32B Q4_K_M 23 GB CPU Offload ~9 tok/s
Qwen 3 32B 32B Q4_K_M 23 GB CPU Offload ~9 tok/s
Command R 35B 35B Q4_K_M 22.5 GB CPU Offload ~9 tok/s
7 model(s) are too large for this hardware.