TEST #2959 RESULTS
03/03/2026 - 3:20 AM
ACCELERATOR
MODEL
228
tokens/s
generation
113
ms
time to first token
11716
tokens/s
prompt
2868
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 5945WX 12-Cores (znver3)
RAM
127.9GB
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
14646
tokens/s
235
tokens/s
74
ms
pp4096+tg256
10344
tokens/s
213
tokens/s
401
ms
pp2048+tg256
13313
tokens/s
233
tokens/s
158
ms
pp2048+tg768
13264
tokens/s
213
tokens/s
159
ms
pp1024+tg1024
15485
tokens/s
232
tokens/s
70
ms
pp1280+tg3072
14504
tokens/s
214
tokens/s
93
ms
pp384+tg1152
16486
tokens/s
235
tokens/s
27
ms
pp64+tg1024
6613
tokens/s
240
tokens/s
14
ms
pp16+tg1536
790
tokens/s
240
tokens/s
24
ms