TEST #1507 RESULTS
09/11/2025 - 10:12 PM
ACCELERATOR
MODEL
101
tokens/s
generation
368
ms
time to first token
3454
tokens/s
prompt
982
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 3945WX 12-Cores (znver2)
RAM
125.6GB
OS
Linux
Kernel Release
6.12.46-1-lts
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Tue, 09 Sep 2025 18:30:25 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3609
tokens/s
105
tokens/s
293
ms
pp4096+tg256
3395
tokens/s
93.1
tokens/s
1.22
sec
pp2048+tg256
3929
tokens/s
102
tokens/s
531
ms
pp2048+tg768
3933
tokens/s
101
tokens/s
531
ms
pp1024+tg1024
4237
tokens/s
104
tokens/s
251
ms
pp1280+tg3072
4077
tokens/s
95.0
tokens/s
323
ms
pp384+tg1152
4090
tokens/s
101
tokens/s
103
ms
pp64+tg1024
2974
tokens/s
104
tokens/s
30
ms
pp16+tg1536
837
tokens/s
102
tokens/s
28
ms