TEST #3694 RESULTS
06/06/2026 - 10:41 AM
ACCELERATOR
MODEL
66.3
tokens/s
generation
419
ms
time to first token
3075
tokens/s
prompt
786
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 9 275HX (arrowlake-s)
RAM
15.5GB
OS
Linux
Kernel Release
6.6.114.1-microsoft-standard-WSL2
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Mon Dec 1 20:46:23 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3559
tokens/s
61.4
tokens/s
304
ms
pp4096+tg256
2979
tokens/s
64.7
tokens/s
1.39
sec
pp2048+tg256
3549
tokens/s
68.9
tokens/s
601
ms
pp2048+tg768
3303
tokens/s
64.4
tokens/s
636
ms
pp1024+tg1024
3585
tokens/s
70.8
tokens/s
301
ms
pp1280+tg3072
3832
tokens/s
62.9
tokens/s
349
ms
pp384+tg1152
3820
tokens/s
67.6
tokens/s
117
ms
pp64+tg1024
2263
tokens/s
67.9
tokens/s
42
ms
pp16+tg1536
781
tokens/s
67.7
tokens/s
34
ms