Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
LLM Decoding Throughput on Apple M4 Pro tg128
Loading...
465
Throughput (tokens/s)
BaseRT
52.12
159.31
266.5
373.69
Jul 1, 2026
Throughput (tokens/s)
Throughput Ratio
Updated 24d ago
Evaluation Results
Method
Method
Links
Throughput (tokens/s)
Throughput Ratio
BaseRT
Model=Qwen3 0.6B, Quan...
2026.07
465
1.17
uzu
Model=Qwen3 0.6B, Quan...
2026.07
398
-
BaseRT
Model=Qwen3 0.6B, Quan...
2026.07
321
1.19
BaseRT
Model=Llama 3.2 1B, Qu...
2026.07
295
1.02
uzu
Model=Llama 3.2 1B, Qu...
2026.07
290
-
uzu
Model=Qwen3 0.6B, Quan...
2026.07
271
-
BaseRT
Model=Llama 3.2 1B, Qu...
2026.07
184
1.06
uzu
Model=Llama 3.2 1B, Qu...
2026.07
173
-
uzu
Model=Llama 3.2 3B, Qu...
2026.07
119
-
BaseRT
Model=Llama 3.2 3B, Qu...
2026.07
117
0.99
BaseRT
Model=Llama 3.2 3B, Qu...
2026.07
71
1.04
uzu
Model=Llama 3.2 3B, Qu...
2026.07
68
-
Feedback
Search any
task
Search any
task
