TEST #4185 RESULTS
09/18/2026 - 6:26 PM
ACCELERATOR
MODEL
47.3
tokens/s
generation
904
ms
time to first token
1426
tokens/s
prompt
421
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i5-12400F (alderlake)
RAM
31.8GB
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
1514
tokens/s
41.2
tokens/s
696
ms
pp4096+tg256
1357
tokens/s
44.4
tokens/s
3.04
sec
pp2048+tg256
1605
tokens/s
49.0
tokens/s
1.30
sec
pp2048+tg768
1600
tokens/s
48.3
tokens/s
1.30
sec
pp1024+tg1024
1742
tokens/s
50.3
tokens/s
607
ms
pp1280+tg3072
1681
tokens/s
45.2
tokens/s
782
ms
pp384+tg1152
1453
tokens/s
46.3
tokens/s
292
ms
pp64+tg1024
1447
tokens/s
51.3
tokens/s
63
ms
pp16+tg1536
431
tokens/s
49.2
tokens/s
56
ms