TEST #39 RESULTS
03/28/2025 - 6:16 PM
ACCELERATOR
MODEL
84.0
tokens/s
generation
621
ms
time to first token
2494
tokens/s
prompt
696
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
127.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3051
tokens/s
89.8
tokens/s
346
ms
pp4096+tg256
1660
tokens/s
48.8
tokens/s
2.49
sec
pp2048+tg256
2371
tokens/s
68.9
tokens/s
878
ms
pp2048+tg768
2295
tokens/s
67.6
tokens/s
906
ms
pp1024+tg1024
3031
tokens/s
84.8
tokens/s
348
ms
pp1280+tg3072
2870
tokens/s
60.0
tokens/s
457
ms
pp384+tg1152
3838
tokens/s
102
tokens/s
108
ms
pp64+tg1024
1850
tokens/s
122
tokens/s
41
ms
pp16+tg1536
1481
tokens/s
112
tokens/s
17
ms