TEST #316 RESULTS
04/05/2025 - 11:50 PM
ACCELERATOR
MODEL
45.5
tokens/s
generation
382
ms
time to first token
3371
tokens/s
prompt
738
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 7950X3D 16-Core Processor (znver4)
RAM
61.9GB
OS
Linux
Kernel Release
6.13.9-200.fc41.x86_64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Sat Mar 29 01:29:31 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4205
tokens/s
49.8
tokens/s
264
ms
pp4096+tg256
3331
tokens/s
37.2
tokens/s
1.26
sec
pp2048+tg256
3881
tokens/s
44.0
tokens/s
550
ms
pp2048+tg768
3877
tokens/s
42.6
tokens/s
551
ms
pp1024+tg1024
4197
tokens/s
46.1
tokens/s
265
ms
pp1280+tg3072
3990
tokens/s
41.1
tokens/s
342
ms
pp384+tg1152
4365
tokens/s
48.6
tokens/s
109
ms
pp64+tg1024
1922
tokens/s
50.3
tokens/s
52
ms
pp16+tg1536
567
tokens/s
49.4
tokens/s
48
ms