TEST #2960 RESULTS
03/03/2026 - 3:23 AM
ACCELERATOR
MODEL
79.0
tokens/s
generation
411
ms
time to first token
3125
tokens/s
prompt
844
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 5945WX 12-Cores (znver3)
RAM
127.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3785
tokens/s
81.7
tokens/s
283
ms
pp4096+tg256
3012
tokens/s
74.0
tokens/s
1.37
sec
pp2048+tg256
3478
tokens/s
79.9
tokens/s
602
ms
pp2048+tg768
3493
tokens/s
77.8
tokens/s
599
ms
pp1024+tg1024
3751
tokens/s
79.8
tokens/s
285
ms
pp1280+tg3072
3631
tokens/s
76.6
tokens/s
365
ms
pp384+tg1152
3755
tokens/s
79.9
tokens/s
115
ms
pp64+tg1024
2549
tokens/s
81.2
tokens/s
37
ms
pp16+tg1536
673
tokens/s
80.5
tokens/s
36
ms