Apple M3 Ultra 20P+8E+60GPU
GPU
96
GB
PERFORMANCE OVERVIEW
Model
Llama 3.2 1B Instruct
Q4_K - Medium1.5B
Qwen2.5 14B Instruct
Q4_K - Medium14.8B
Prompt Speed
4394tokens/s
445tokens/s
Generation Speed
173tokens/s
34.4tokens/s
Time to First Token
261ms
2.67sec
LocalScore
1428
179
COMPARE MODELS
2 models tested
Select Models
Llama 3.2 1B Instruct
Q4_K - Medium
Qwen2.5 14B Instruct
Q4_K - Medium