TEST #381 RESULTS
04/06/2025 - 3:18 PM
ACCELERATOR
MODEL
2.8
tokens/s
generation
23.14
sec
time to first token
66
tokens/s
prompt
20
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-10750H CPU @ 2.60GHz (skylake)
RAM
31.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
56
tokens/s
4.0
tokens/s
18.62
sec
pp4096+tg256
42
tokens/s
1.7
tokens/s
98.33
sec
pp2048+tg256
77
tokens/s
2.6
tokens/s
26.98
sec
pp2048+tg768
77
tokens/s
2.4
tokens/s
27.12
sec
pp1024+tg1024
81
tokens/s
2.9
tokens/s
13.00
sec
pp1280+tg3072
78
tokens/s
2.1
tokens/s
16.84
sec
pp384+tg1152
74
tokens/s
3.0
tokens/s
5.47
sec
pp64+tg1024
76
tokens/s
3.3
tokens/s
1.11
sec
pp16+tg1536
31
tokens/s
3.1
tokens/s
778
ms