TEST #2671 RESULTS
02/15/2026 - 5:45 PM
ACCELERATOR
MODEL
178
tokens/s
generation
168
ms
time to first token
8269
tokens/s
prompt
2065
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600X 6-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
10026
tokens/s
147
tokens/s
111
ms
pp4096+tg256
6740
tokens/s
84.7
tokens/s
626
ms
pp2048+tg256
8930
tokens/s
197
tokens/s
235
ms
pp2048+tg768
9081
tokens/s
198
tokens/s
231
ms
pp1024+tg1024
10637
tokens/s
199
tokens/s
101
ms
pp1280+tg3072
9908
tokens/s
193
tokens/s
134
ms
pp384+tg1152
11955
tokens/s
196
tokens/s
37
ms
pp64+tg1024
5840
tokens/s
195
tokens/s
16
ms
pp16+tg1536
1307
tokens/s
195
tokens/s
18
ms