TEST #358 RESULTS
04/06/2025 - 6:13 AM
ACCELERATOR
MODEL
162
tokens/s
generation
126
ms
time to first token
10791
tokens/s
prompt
2402
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 3955WX 16-Cores (znver2)
RAM
127.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
13447
tokens/s
156
tokens/s
83
ms
pp4096+tg256
9306
tokens/s
156
tokens/s
446
ms
pp2048+tg256
11854
tokens/s
163
tokens/s
178
ms
pp2048+tg768
11974
tokens/s
166
tokens/s
177
ms
pp1024+tg1024
14378
tokens/s
164
tokens/s
77
ms
pp1280+tg3072
13179
tokens/s
162
tokens/s
103
ms
pp384+tg1152
14847
tokens/s
161
tokens/s
33
ms
pp64+tg1024
6878
tokens/s
165
tokens/s
18
ms
pp16+tg1536
1259
tokens/s
165
tokens/s
19
ms