TEST #569 RESULTS
04/08/2025 - 9:05 AM
ACCELERATOR
MODEL
74.0
tokens/s
generation
1.05
sec
time to first token
1677
tokens/s
prompt
491
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
QEMU Virtual CPU version 2.5+
RAM
295GB
OS
Linux
Kernel Release
6.8.0-57-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #59-Ubuntu SMP PREEMPT_DYNAMIC Sat Mar 15 17:40:59 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
358
tokens/s
72.0
tokens/s
2.87
sec
pp4096+tg256
1199
tokens/s
62.7
tokens/s
3.43
sec
pp2048+tg256
2223
tokens/s
72.6
tokens/s
935
ms
pp2048+tg768
2318
tokens/s
70.8
tokens/s
898
ms
pp1024+tg1024
2290
tokens/s
76.2
tokens/s
460
ms
pp1280+tg3072
2526
tokens/s
68.0
tokens/s
520
ms
pp384+tg1152
2204
tokens/s
79.9
tokens/s
186
ms
pp64+tg1024
1619
tokens/s
82.9
tokens/s
51
ms
pp16+tg1536
358
tokens/s
81.0
tokens/s
57
ms