TEST #3454 RESULTS
05/06/2026 - 7:42 AM
ACCELERATOR
MODEL
47.4
tokens/s
generation
4.15
sec
time to first token
362
tokens/s
prompt
160
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
31.9GB
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
405
tokens/s
49.3
tokens/s
2.55
sec
pp4096+tg256
265
tokens/s
40.6
tokens/s
15.47
sec
pp2048+tg256
329
tokens/s
45.9
tokens/s
6.24
sec
pp2048+tg768
334
tokens/s
45.1
tokens/s
6.17
sec
pp1024+tg1024
406
tokens/s
48.4
tokens/s
2.54
sec
pp1280+tg3072
403
tokens/s
44.0
tokens/s
3.20
sec
pp384+tg1152
422
tokens/s
50.4
tokens/s
929
ms
pp64+tg1024
388
tokens/s
51.7
tokens/s
184
ms
pp16+tg1536
303
tokens/s
51.0
tokens/s
72
ms