TEST #451 RESULTS
04/07/2025 - 12:43 AM
ACCELERATOR
MODEL
188
tokens/s
generation
168
ms
time to first token
9571
tokens/s
prompt
2205
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7713P 64-Core Processor (znver3)
RAM
125.4GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 15 19:18:46 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
12196
tokens/s
185
tokens/s
90
ms
pp4096+tg256
5991
tokens/s
166
tokens/s
690
ms
pp2048+tg256
8920
tokens/s
186
tokens/s
235
ms
pp2048+tg768
8977
tokens/s
184
tokens/s
234
ms
pp1024+tg1024
12279
tokens/s
192
tokens/s
89
ms
pp1280+tg3072
11389
tokens/s
178
tokens/s
118
ms
pp384+tg1152
16628
tokens/s
197
tokens/s
28
ms
pp64+tg1024
7506
tokens/s
202
tokens/s
13
ms
pp16+tg1536
2251
tokens/s
198
tokens/s
12
ms