TEST #3367 RESULTS
04/23/2026 - 3:54 AM
ACCELERATOR
MODEL
78.2
tokens/s
generation
521
ms
time to first token
2512
tokens/s
prompt
723
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-14500 (alderlake)
RAM
62.6GB
OS
Linux
Kernel Release
6.12.73+deb13-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.12.73-1 (2026-02-17)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2928
tokens/s
82.5
tokens/s
362
ms
pp4096+tg256
2367
tokens/s
70.9
tokens/s
1.74
sec
pp2048+tg256
2720
tokens/s
78.2
tokens/s
766
ms
pp2048+tg768
2714
tokens/s
76.8
tokens/s
767
ms
pp1024+tg1024
2904
tokens/s
80.0
tokens/s
365
ms
pp1280+tg3072
2797
tokens/s
74.1
tokens/s
470
ms
pp384+tg1152
2958
tokens/s
80.1
tokens/s
142
ms
pp64+tg1024
2491
tokens/s
81.0
tokens/s
37
ms
pp16+tg1536
734
tokens/s
80.4
tokens/s
33
ms