TEST #3388 RESULTS
04/26/2026 - 3:25 PM
ACCELERATOR
MODEL
31.1
tokens/s
generation
3.14
sec
time to first token
372
tokens/s
prompt
154
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2680 v4 @ 2.40GHz (broadwell)
RAM
31.2GB
OS
Linux
Kernel Release
6.17.13-3-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 6.17.13-3 (2026-03-31T21:50Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
454
tokens/s
33.1
tokens/s
2.29
sec
pp4096+tg256
425
tokens/s
24.8
tokens/s
9.69
sec
pp2048+tg256
438
tokens/s
29.7
tokens/s
4.71
sec
pp2048+tg768
436
tokens/s
29.0
tokens/s
4.73
sec
pp1024+tg1024
446
tokens/s
31.8
tokens/s
2.33
sec
pp1280+tg3072
431
tokens/s
28.0
tokens/s
3.00
sec
pp384+tg1152
437
tokens/s
33.7
tokens/s
908
ms
pp64+tg1024
209
tokens/s
35.2
tokens/s
333
ms
pp16+tg1536
69
tokens/s
34.4
tokens/s
260
ms