TEST #2955 RESULTS
03/02/2026 - 12:28 PM
ACCELERATOR
MODEL
80.1
tokens/s
generation
265
ms
time to first token
5033
tokens/s
prompt
1151
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
6294
tokens/s
79.6
tokens/s
175
ms
pp4096+tg256
4534
tokens/s
73.9
tokens/s
918
ms
pp2048+tg256
5567
tokens/s
78.5
tokens/s
380
ms
pp2048+tg768
5543
tokens/s
79.3
tokens/s
383
ms
pp1024+tg1024
6256
tokens/s
81.3
tokens/s
176
ms
pp1280+tg3072
6194
tokens/s
78.5
tokens/s
219
ms
pp384+tg1152
6633
tokens/s
82.3
tokens/s
70
ms
pp64+tg1024
3251
tokens/s
83.9
tokens/s
32
ms
pp16+tg1536
1026
tokens/s
83.4
tokens/s
28
ms