TEST #3295 RESULTS
04/12/2026 - 2:41 AM
ACCELERATOR
MODEL
45.3
tokens/s
generation
155
ms
time to first token
8635
tokens/s
prompt
1361
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
31.1GB
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
11386
tokens/s
46.1
tokens/s
111
ms
pp4096+tg256
9502
tokens/s
44.1
tokens/s
454
ms
pp2048+tg256
11009
tokens/s
45.8
tokens/s
208
ms
pp2048+tg768
11097
tokens/s
45.1
tokens/s
207
ms
pp1024+tg1024
11267
tokens/s
45.2
tokens/s
115
ms
pp1280+tg3072
10444
tokens/s
44.7
tokens/s
145
ms
pp384+tg1152
10175
tokens/s
45.4
tokens/s
62
ms
pp64+tg1024
2181
tokens/s
45.5
tokens/s
49
ms
pp16+tg1536
653
tokens/s
45.5
tokens/s
44
ms