TEST #3279 RESULTS
04/10/2026 - 9:15 PM
ACCELERATOR
MODEL
31.7
tokens/s
generation
1.11
sec
time to first token
1217
tokens/s
prompt
326
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i5-12500H (alderlake)
RAM
15.7GB
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
1425
tokens/s
32.9
tokens/s
749
ms
pp4096+tg256
1085
tokens/s
29.3
tokens/s
3.81
sec
pp2048+tg256
1283
tokens/s
31.5
tokens/s
1.63
sec
pp2048+tg768
1278
tokens/s
31.1
tokens/s
1.63
sec
pp1024+tg1024
1384
tokens/s
32.0
tokens/s
771
ms
pp1280+tg3072
1351
tokens/s
30.4
tokens/s
978
ms
pp384+tg1152
1518
tokens/s
32.5
tokens/s
283
ms
pp64+tg1024
1215
tokens/s
32.8
tokens/s
82
ms
pp16+tg1536
414
tokens/s
32.6
tokens/s
69
ms