TEST #3589 RESULTS
05/21/2026 - 4:49 PM
ACCELERATOR
MODEL
216
tokens/s
generation
108
ms
time to first token
12560
tokens/s
prompt
2931
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
16196
tokens/s
214
tokens/s
68
ms
pp4096+tg256
11790
tokens/s
205
tokens/s
353
ms
pp2048+tg256
14898
tokens/s
216
tokens/s
143
ms
pp2048+tg768
14763
tokens/s
218
tokens/s
143
ms
pp1024+tg1024
17332
tokens/s
219
tokens/s
64
ms
pp1280+tg3072
16129
tokens/s
211
tokens/s
84
ms
pp384+tg1152
17699
tokens/s
222
tokens/s
26
ms
pp64+tg1024
3977
tokens/s
220
tokens/s
21
ms
pp16+tg1536
256
tokens/s
219
tokens/s
67
ms