TEST #3551 RESULTS
05/16/2026 - 9:01 AM
ACCELERATOR
MODEL
75.5
tokens/s
generation
444
ms
time to first token
2891
tokens/s
prompt
789
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i7-13700KF (alderlake)
RAM
63.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3502
tokens/s
77.0
tokens/s
306
ms
pp4096+tg256
2802
tokens/s
70.3
tokens/s
1.48
sec
pp2048+tg256
3200
tokens/s
76.7
tokens/s
654
ms
pp2048+tg768
3210
tokens/s
75.4
tokens/s
652
ms
pp1024+tg1024
3473
tokens/s
76.0
tokens/s
308
ms
pp1280+tg3072
3342
tokens/s
73.6
tokens/s
396
ms
pp384+tg1152
3356
tokens/s
75.8
tokens/s
127
ms
pp64+tg1024
2434
tokens/s
77.2
tokens/s
39
ms
pp16+tg1536
698
tokens/s
77.1
tokens/s
36
ms