TEST #3046 RESULTS
03/08/2026 - 1:51 PM
ACCELERATOR
MODEL
88.0
tokens/s
generation
535
ms
time to first token
2204
tokens/s
prompt
713
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 3600 6-Core Processor (znver2)
RAM
31.2GB
OS
Linux
Kernel Release
6.8.0-101-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #101-Ubuntu SMP PREEMPT_DYNAMIC Mon Feb 9 10:15:05 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2867
tokens/s
94.5
tokens/s
368
ms
pp4096+tg256
2401
tokens/s
70.8
tokens/s
1.72
sec
pp2048+tg256
2702
tokens/s
84.5
tokens/s
770
ms
pp2048+tg768
2699
tokens/s
82.4
tokens/s
771
ms
pp1024+tg1024
2883
tokens/s
90.6
tokens/s
366
ms
pp1280+tg3072
2602
tokens/s
79.2
tokens/s
503
ms
pp384+tg1152
2551
tokens/s
94.9
tokens/s
161
ms
pp64+tg1024
617
tokens/s
98.5
tokens/s
114
ms
pp16+tg1536
516
tokens/s
96.3
tokens/s
41
ms