TEST #351 RESULTS
04/06/2025 - 3:50 AM
ACCELERATOR
MODEL
30.7
tokens/s
generation
617
ms
time to first token
2112
tokens/s
prompt
472
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
31GB
OS
Linux
Kernel Release
6.13.8-201.nobara.fc41.x86_64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Tue Mar 25 08:23:44 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2620
tokens/s
34.4
tokens/s
420
ms
pp4096+tg256
2036
tokens/s
25.2
tokens/s
2.05
sec
pp2048+tg256
2336
tokens/s
29.8
tokens/s
910
ms
pp2048+tg768
2402
tokens/s
28.0
tokens/s
886
ms
pp1024+tg1024
2501
tokens/s
30.4
tokens/s
440
ms
pp1280+tg3072
2525
tokens/s
27.3
tokens/s
538
ms
pp384+tg1152
2755
tokens/s
32.7
tokens/s
168
ms
pp64+tg1024
1413
tokens/s
34.9
tokens/s
72
ms
pp16+tg1536
418
tokens/s
33.3
tokens/s
66
ms