Local LLM Performance on Apple Silicon

Detailed per-chip benchmark pages for 109 model+chip combinations. Find out exactly how fast any model runs on your Mac — from M1 with 8 GB RAM to M5 Max with 128 GB.

These pages answer how fast is this model on that chip. For the inverse — what should I run on the Mac I already own — go to Best local LLM for every Mac →
Shopping for the machine instead? The Mac Advisor turns a budget and a use case into one config → · Every Mac compared →

Gemma 4 E2B
2.3B • up to 261 tok/s
1 chips
M5 Max 261 tok/s 128GB
Qwen 3.5 4B
4B • up to 161 tok/s
1 chips
M5 Max 161 tok/s 64GB
Gemma 4 E4B
4B • up to 161 tok/s
4 chips
M5 Max 161 tok/s 128GB M5 Pro 91 tok/s 24GB M4 Pro 82 tok/s 24GB M6 54 tok/s 16GB
Qwen3-Coder-Next
80B • up to 130 tok/s
2 chips
M5 Ultra 130 tok/s 96GB M5 Max 84 tok/s 64GB
Qwen 3.6-35B-A3B
35B • up to 120 tok/s
2 chips
M4 Max 120 tok/s 48GB M6 64 tok/s 32GB
KAT-Coder-V2.5
35B-A3B • up to 117 tok/s
2 chips
M5 Max 117 tok/s 64GB M6 56 tok/s 32GB
Gemma 4 26B-A4B
26B • up to 106 tok/s
3 chips
M5 Max 106 tok/s 128GB M4 Pro 67 tok/s 24GB M6 48 tok/s 32GB
GPT-oss 20B
21B-A3.6B • up to 103 tok/s
5 chips
M5 Max 103 tok/s 64GB M4 Max 97 tok/s 64GB M4 Pro 64 tok/s 24GB M6 45 tok/s 32GB M4 34 tok/s 16GB
GPT-oss 120B
117B-A5.1B • up to 97 tok/s
2 chips
M5 Ultra 97 tok/s 96GB M5 Max 58 tok/s 128GB
Qwen 3 8B
8B • up to 91 tok/s
4 chips
M5 Max 91 tok/s 128GB M4 Pro 44 tok/s 24GB M2 17 tok/s 16GB M1 11 tok/s 16GB
Llama 3.1 8B
8B • up to 91 tok/s
3 chips
M5 Max 91 tok/s 128GB M4 20 tok/s 16GB M1 11 tok/s 16GB
Qwen3.8-Flash-Next
125B-A6B • up to 84 tok/s
4 chips
M5 Ultra 84 tok/s 256GB M3 Ultra 62 tok/s 256GB M5 Max 49 tok/s 128GB M4 Max 44 tok/s 128GB
Qwen 3.5 9B
9B • up to 82 tok/s
6 chips
M5 Max 82 tok/s 64GB M4 Pro 39 tok/s 24GB M6 25 tok/s 16GB M4 18 tok/s 16GB M3 15 tok/s 16GB M1 10 tok/s 16GB
Qwen 3.5 122B-A10B
122B-A10B • up to 80 tok/s
5 chips
M5 Ultra 80 tok/s 256GB M3 Ultra 60 tok/s 96GB M5 Max 47 tok/s 128GB M4 Max 42 tok/s 128GB M3 Max 32 tok/s 128GB
LFM2.5-2.6B
2.6B • up to 79 tok/s
1 chips
M6 79 tok/s 16GB
Mistral Small 4
119B • up to 79 tok/s
1 chips
M5 Ultra 79 tok/s 96GB
Laguna S 2.1
118B • up to 68 tok/s
1 chips
M5 Ultra 68 tok/s 96GB
Gemma 3 12B
12B • up to 64 tok/s
3 chips
M5 Max 64 tok/s 64GB M3 11 tok/s 16GB M2 11 tok/s 16GB
Gemma 4 12B
12B • up to 64 tok/s
5 chips
M5 Max 64 tok/s 64GB M4 Max 57 tok/s 64GB M4 Pro 30 tok/s 24GB M6 19 tok/s 32GB M4 13 tok/s 16GB
MiniMax M2.5
230B-A10B • up to 57 tok/s
2 chips
M5 Ultra 57 tok/s 256GB M3 Ultra 41 tok/s 256GB
Qwen3.8-27B
27.8B • up to 55 tok/s
7 chips
M5 Ultra 55 tok/s 256GB M3 Ultra 40 tok/s 96GB M5 Max 29 tok/s 128GB M4 Max 26 tok/s 64GB M3 Max 19 tok/s 64GB M4 Pro 13 tok/s 48GB M6 8 tok/s 24GB
Inkling-Small
276B • up to 49 tok/s
1 chips
M5 Ultra 49 tok/s 256GB
GLM-4.5-Air
106B • up to 49 tok/s
1 chips
M5 Ultra 49 tok/s 96GB
Gemma 3 4B
4B • up to 48 tok/s
1 chips
M3 Pro 48 tok/s 18GB
DeepSeek V4 Flash Vision-Exp
305B-A13B • up to 46 tok/s
3 chips
M5 Ultra 46 tok/s 256GB M5 Max 39 tok/s 128GB M3 Ultra 32 tok/s 256GB
DeepSeek V4 Flash
284B-A13B • up to 45 tok/s
4 chips
M5 Ultra 45 tok/s 256GB M5 Max 39 tok/s 128GB M3 Ultra 32 tok/s 256GB M3 Max 27 tok/s 128GB
Nemotron 3.5 Lightning
30B-A3B • up to 45 tok/s
1 chips
M6 45 tok/s 32GB
MiMo-V2.6-Flash
309B-A15B • up to 42 tok/s
2 chips
M5 Ultra 42 tok/s 256GB M3 Ultra 30 tok/s 256GB
Qwen 3.6-27B
27B • up to 38 tok/s
6 chips
M1 Ultra 38 tok/s 64GB M5 Max 30 tok/s 64GB M4 Max 27 tok/s 64GB M2 Max 20 tok/s 32GB M5 Pro 15 tok/s 24GB M6 9 tok/s 24GB
GLM-5.3-Flash
320B-A18B • up to 35 tok/s
3 chips
M5 Ultra 35 tok/s 256GB M5 Max 27 tok/s 128GB M3 Ultra 24 tok/s 256GB
Qwen3-235B-A22B
235B • up to 29 tok/s
1 chips
M5 Ultra 29 tok/s 256GB
DeepSeek R1 8B
8B • up to 28 tok/s
1 chips
M6 28 tok/s 16GB
Gemma 3 27B
27B • up to 27 tok/s
2 chips
M4 Max 27 tok/s 48GB M4 Pro 14 tok/s 24GB
Muse Glimmer 30B
30B • up to 27 tok/s
2 chips
M5 Max 27 tok/s 64GB M6 9 tok/s 32GB
Gemma 4 31B
31B • up to 26 tok/s
3 chips
M5 Max 26 tok/s 128GB M4 Pro 12 tok/s 24GB M6 8 tok/s 32GB
Qwen 3 14B
14B • up to 26 tok/s
2 chips
M4 Pro 26 tok/s 24GB M3 10 tok/s 16GB
Qwen 3 32B
32B • up to 23 tok/s
1 chips
M4 Max 23 tok/s 64GB
DeepSeek R1 70B
70B • up to 23 tok/s
2 chips
M5 Ultra 23 tok/s 96GB M5 Max 12 tok/s 128GB
GLM-4.7-Flash
31B • up to 23 tok/s
1 chips
M4 Max 23 tok/s 48GB
Llama 3.3 70B
70B • up to 23 tok/s
1 chips
M5 Ultra 23 tok/s 96GB
Qwen 2.5 72B
72B • up to 23 tok/s
1 chips
M5 Ultra 23 tok/s 96GB
GLM-5.3
753B-A40B • up to 17 tok/s
2 chips
M5 Ultra 17 tok/s 512GB M3 Ultra 11 tok/s 512GB
Phi-4 14B
14B • up to 16 tok/s
2 chips
M6 16 tok/s 16GB M4 12 tok/s 16GB
GLM 5.2
753B-A40B • up to 15 tok/s
1 chips
M3 Ultra 15 tok/s 512GB
Llama 3.1 405B
405B • up to 4 tok/s
1 chips
M5 Ultra 4 tok/s 512GB