TEST #2952 RESULTS
03/02/2026 - 12:19 PM
ACCELERATOR
MODEL
233
tokens/s
generation
117
ms
time to first token
11173
tokens/s
prompt
2816
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
13944
tokens/s
213
tokens/s
79
ms
pp4096+tg256
9988
tokens/s
198
tokens/s
416
ms
pp2048+tg256
13165
tokens/s
228
tokens/s
160
ms
pp2048+tg768
13351
tokens/s
223
tokens/s
158
ms
pp1024+tg1024
15426
tokens/s
235
tokens/s
71
ms
pp1280+tg3072
14165
tokens/s
232
tokens/s
95
ms
pp384+tg1152
16374
tokens/s
256
tokens/s
27
ms
pp64+tg1024
3257
tokens/s
261
tokens/s
24
ms
pp16+tg1536
882
tokens/s
256
tokens/s
22
ms