TEST #768 RESULTS
04/25/2025 - 9:55 PM
ACCELERATOR
MODEL
34.8
tokens/s
generation
1.15
sec
time to first token
1050
tokens/s
prompt
316
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i7-11800H @ 2.30GHz (tigerlake)
RAM
15.8GB
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
1508
tokens/s
44.9
tokens/s
701
ms
pp4096+tg256
1119
tokens/s
35.6
tokens/s
3.70
sec
pp2048+tg256
1345
tokens/s
42.3
tokens/s
1.55
sec
pp2048+tg768
1337
tokens/s
41.6
tokens/s
1.55
sec
pp1024+tg1024
1264
tokens/s
39.3
tokens/s
851
ms
pp1280+tg3072
944
tokens/s
28.2
tokens/s
1.38
sec
pp384+tg1152
946
tokens/s
25.7
tokens/s
437
ms
pp64+tg1024
653
tokens/s
24.9
tokens/s
132
ms
pp16+tg1536
334
tokens/s
30.3
tokens/s
76
ms