TEST #2082 RESULTS
01/03/2026 - 4:19 PM
ACCELERATOR
MODEL
22.5
tokens/s
generation
1.39
sec
time to first token
984
tokens/s
prompt
251
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 7 265H (arrowlake)
RAM
30.3GB
OS
Linux
Kernel Release
6.17.0-8-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #8-Ubuntu SMP PREEMPT_DYNAMIC Fri Nov 14 21:44:46 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1097
tokens/s
23.5
tokens/s
958
ms
pp4096+tg256
875
tokens/s
22.1
tokens/s
4.73
sec
pp2048+tg256
1007
tokens/s
23.0
tokens/s
2.06
sec
pp2048+tg768
1012
tokens/s
22.2
tokens/s
2.05
sec
pp1024+tg1024
1107
tokens/s
22.6
tokens/s
951
ms
pp1280+tg3072
1039
tokens/s
21.3
tokens/s
1.26
sec
pp384+tg1152
1218
tokens/s
22.3
tokens/s
339
ms
pp64+tg1024
967
tokens/s
22.7
tokens/s
95
ms
pp16+tg1536
533
tokens/s
22.4
tokens/s
53
ms