TEST #78 RESULTS
03/28/2025 - 9:11 PM
ACCELERATOR
MODEL
83.9
tokens/s
generation
99
ms
time to first token
13164
tokens/s
prompt
2232
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 9374F 32-Core Processor (znver4)
RAM
1448.6GB
OS
Linux
Kernel Release
6.8.0-51-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #52~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Mon Dec 9 15:00:52 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
17916
tokens/s
87.2
tokens/s
69
ms
pp4096+tg256
14530
tokens/s
77.4
tokens/s
300
ms
pp2048+tg256
16741
tokens/s
82.0
tokens/s
135
ms
pp2048+tg768
17260
tokens/s
81.3
tokens/s
131
ms
pp1024+tg1024
18017
tokens/s
82.5
tokens/s
69
ms
pp1280+tg3072
14723
tokens/s
80.2
tokens/s
99
ms
pp384+tg1152
15052
tokens/s
85.2
tokens/s
36
ms
pp64+tg1024
3306
tokens/s
90.5
tokens/s
29
ms
pp16+tg1536
932
tokens/s
88.8
tokens/s
27
ms