TEST #3590 RESULTS
05/21/2026 - 6:02 PM
ACCELERATOR
MODEL
208
tokens/s
generation
108
ms
time to first token
12930
tokens/s
prompt
2915
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
17456
tokens/s
210
tokens/s
64
ms
pp4096+tg256
11813
tokens/s
197
tokens/s
353
ms
pp2048+tg256
15132
tokens/s
206
tokens/s
141
ms
pp2048+tg768
15014
tokens/s
203
tokens/s
142
ms
pp1024+tg1024
17292
tokens/s
204
tokens/s
65
ms
pp1280+tg3072
15857
tokens/s
204
tokens/s
86
ms
pp384+tg1152
18067
tokens/s
214
tokens/s
26
ms
pp64+tg1024
5527
tokens/s
213
tokens/s
17
ms
pp16+tg1536
207
tokens/s
219
tokens/s
82
ms