TEST #2757 RESULTS
02/19/2026 - 7:07 AM
ACCELERATOR
MODEL
37.1
tokens/s
generation
657
ms
time to first token
2110
tokens/s
prompt
492
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.6GB
OS
Linux
Kernel Release
6.17.0-1008-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #8-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 21 17:56:56 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2582
tokens/s
37.1
tokens/s
423
ms
pp4096+tg256
1762
tokens/s
33.6
tokens/s
2.36
sec
pp2048+tg256
2228
tokens/s
36.5
tokens/s
947
ms
pp2048+tg768
2230
tokens/s
36.5
tokens/s
946
ms
pp1024+tg1024
2583
tokens/s
37.7
tokens/s
423
ms
pp1280+tg3072
2480
tokens/s
35.7
tokens/s
544
ms
pp384+tg1152
3041
tokens/s
38.5
tokens/s
153
ms
pp64+tg1024
1601
tokens/s
39.1
tokens/s
67
ms
pp16+tg1536
486
tokens/s
38.7
tokens/s
59
ms