TEST #1556 RESULTS
09/24/2025 - 9:00 PM
ACCELERATOR
MODEL
5.5
tokens/s
generation
4.06
sec
time to first token
317
tokens/s
prompt
75
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 7600X 6-Core Processor (znver4)
RAM
27GB
OS
Linux
Kernel Release
6.16.8-200.nobara.fc42.x86_64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Sat Sep 20 08:17:02 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
396
tokens/s
5.9
tokens/s
2.75
sec
pp4096+tg256
316
tokens/s
1.4
tokens/s
13.79
sec
pp2048+tg256
371
tokens/s
4.1
tokens/s
5.80
sec
pp2048+tg768
362
tokens/s
3.8
tokens/s
5.92
sec
pp1024+tg1024
383
tokens/s
5.5
tokens/s
2.85
sec
pp1280+tg3072
372
tokens/s
3.0
tokens/s
3.63
sec
pp384+tg1152
383
tokens/s
7.5
tokens/s
1.10
sec
pp64+tg1024
140
tokens/s
9.7
tokens/s
527
ms
pp16+tg1536
130
tokens/s
8.3
tokens/s
185
ms