Apple M5 Pro 5P+10E+16GPU
GPU
48
GB
PERFORMANCE OVERVIEW
Model
Meta Llama 3.1 8B Instruct
Q4_K - Medium8.0B
Qwen2.5 14B Instruct
Q4_K - Medium14.8B
Prompt Speed
341tokens/s
177tokens/s
Generation Speed
43.7tokens/s
22.5tokens/s
Time to First Token
3.67sec
7.06sec
LocalScore
159
83
COMPARE MODELS
2 models tested
Select Models
Meta Llama 3.1 8B Instruct
Q4_K - Medium
Qwen2.5 14B Instruct
Q4_K - Medium