TEST #63 RESULTS
03/28/2025 - 8:54 PM
ACCELERATOR
MODEL
190
tokens/s
generation
174
ms
time to first token
8579
tokens/s
prompt
2110
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7713 64-Core Processor (znver3)
RAM
503.5GB
OS
Linux
Kernel Release
6.5.0-44-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #44~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Jun 18 14:36:16 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
11623
tokens/s
194
tokens/s
93
ms
pp4096+tg256
5902
tokens/s
169
tokens/s
700
ms
pp2048+tg256
8538
tokens/s
187
tokens/s
245
ms
pp2048+tg768
8793
tokens/s
185
tokens/s
238
ms
pp1024+tg1024
12094
tokens/s
194
tokens/s
90
ms
pp1280+tg3072
11173
tokens/s
180
tokens/s
120
ms
pp384+tg1152
10531
tokens/s
198
tokens/s
41
ms
pp64+tg1024
7610
tokens/s
203
tokens/s
13
ms
pp16+tg1536
948
tokens/s
201
tokens/s
21
ms