TEST #84 RESULTS
03/28/2025 - 9:19 PM
ACCELERATOR
MODEL
121
tokens/s
generation
75
ms
time to first token
18255
tokens/s
prompt
3088
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 9354 32-Core Processor (znver4)
RAM
755.6GB
OS
Linux
Kernel Release
6.5.0-45-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #45~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Mon Jul 15 16:40:02 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
26363
tokens/s
128
tokens/s
47
ms
pp4096+tg256
17092
tokens/s
110
tokens/s
249
ms
pp2048+tg256
22179
tokens/s
121
tokens/s
101
ms
pp2048+tg768
22182
tokens/s
118
tokens/s
101
ms
pp1024+tg1024
26217
tokens/s
122
tokens/s
47
ms
pp1280+tg3072
21291
tokens/s
116
tokens/s
68
ms
pp384+tg1152
22218
tokens/s
124
tokens/s
26
ms
pp64+tg1024
5247
tokens/s
125
tokens/s
19
ms
pp16+tg1536
1506
tokens/s
125
tokens/s
18
ms