TEST #3468 RESULTS
05/07/2026 - 11:16 PM
ACCELERATOR
MODEL
34.1
tokens/s
generation
6.91
sec
time to first token
276
tokens/s
prompt
111
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (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
208
tokens/s
10.1
tokens/s
6.17
sec
pp4096+tg256
234
tokens/s
20.8
tokens/s
18.27
sec
pp2048+tg256
254
tokens/s
30.4
tokens/s
9.17
sec
pp2048+tg768
247
tokens/s
31.3
tokens/s
9.40
sec
pp1024+tg1024
208
tokens/s
38.0
tokens/s
6.19
sec
pp1280+tg3072
167
tokens/s
29.5
tokens/s
9.26
sec
pp384+tg1152
662
tokens/s
46.7
tokens/s
599
ms
pp64+tg1024
51
tokens/s
49.4
tokens/s
3.04
sec
pp16+tg1536
453
tokens/s
50.8
tokens/s
51
ms