TEST #1586 RESULTS
10/05/2025 - 9:33 AM
ACCELERATOR
MODEL
57.9
tokens/s
generation
3.70
sec
time to first token
363
tokens/s
prompt
179
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
31.7GB
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
404
tokens/s
62.3
tokens/s
2.55
sec
pp4096+tg256
334
tokens/s
47.4
tokens/s
12.30
sec
pp2048+tg256
396
tokens/s
57.4
tokens/s
5.19
sec
pp2048+tg768
339
tokens/s
57.0
tokens/s
6.06
sec
pp1024+tg1024
370
tokens/s
58.6
tokens/s
2.79
sec
pp1280+tg3072
422
tokens/s
52.9
tokens/s
3.05
sec
pp384+tg1152
376
tokens/s
60.8
tokens/s
1.04
sec
pp64+tg1024
346
tokens/s
62.9
tokens/s
201
ms
pp16+tg1536
282
tokens/s
61.9
tokens/s
72
ms