TEST #2158 RESULTS
01/06/2026 - 1:58 AM
ACCELERATOR
MODEL
33.1
tokens/s
generation
676
ms
time to first token
2044
tokens/s
prompt
464
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
6.14.0-1015-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #15-Ubuntu SMP PREEMPT_DYNAMIC Tue Nov 25 18:02:16 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2530
tokens/s
33.8
tokens/s
435
ms
pp4096+tg256
1731
tokens/s
30.5
tokens/s
2.40
sec
pp2048+tg256
2158
tokens/s
32.7
tokens/s
981
ms
pp2048+tg768
2174
tokens/s
32.1
tokens/s
973
ms
pp1024+tg1024
2512
tokens/s
33.9
tokens/s
439
ms
pp1280+tg3072
2439
tokens/s
32.2
tokens/s
554
ms
pp384+tg1152
2863
tokens/s
33.7
tokens/s
164
ms
pp64+tg1024
1519
tokens/s
34.7
tokens/s
73
ms
pp16+tg1536
471
tokens/s
34.2
tokens/s
64
ms