TEST #3487 RESULTS
05/10/2026 - 1:28 PM
ACCELERATOR
MODEL
53.4
tokens/s
generation
144
ms
time to first token
9413
tokens/s
prompt
1516
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 9600X 6-Core Processor
RAM
30.9GB
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
13465
tokens/s
60.7
tokens/s
92
ms
pp4096+tg256
10612
tokens/s
45.2
tokens/s
409
ms
pp2048+tg256
12916
tokens/s
51.2
tokens/s
176
ms
pp2048+tg768
11441
tokens/s
51.5
tokens/s
217
ms
pp1024+tg1024
13824
tokens/s
57.4
tokens/s
90
ms
pp1280+tg3072
12234
tokens/s
49.8
tokens/s
122
ms
pp384+tg1152
7472
tokens/s
56.9
tokens/s
103
ms
pp64+tg1024
2189
tokens/s
54.5
tokens/s
45
ms
pp16+tg1536
564
tokens/s
53.6
tokens/s
45
ms