TEST #1198 RESULTS
07/08/2025 - 7:28 PM
ACCELERATOR
MODEL
176
tokens/s
generation
104
ms
time to first token
12960
tokens/s
prompt
2801
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper 3970X 32-Core Processor (znver2)
RAM
63.9GB
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
17224
tokens/s
177
tokens/s
65
ms
pp4096+tg256
11399
tokens/s
172
tokens/s
365
ms
pp2048+tg256
14673
tokens/s
174
tokens/s
146
ms
pp2048+tg768
14496
tokens/s
178
tokens/s
147
ms
pp1024+tg1024
16884
tokens/s
178
tokens/s
67
ms
pp1280+tg3072
16065
tokens/s
176
tokens/s
85
ms
pp384+tg1152
17573
tokens/s
177
tokens/s
28
ms
pp64+tg1024
6964
tokens/s
178
tokens/s
15
ms
pp16+tg1536
1360
tokens/s
178
tokens/s
18
ms