TEST #1709 RESULTS
10/28/2025 - 3:55 AM
ACCELERATOR
MODEL
109
tokens/s
generation
80
ms
time to first token
16148
tokens/s
prompt
2799
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
377.4GB
OS
Linux
Kernel Release
6.8.0-85-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #85-Ubuntu SMP PREEMPT_DYNAMIC Thu Sep 18 15:26:59 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
21810
tokens/s
112
tokens/s
56
ms
pp4096+tg256
18287
tokens/s
102
tokens/s
234
ms
pp2048+tg256
20172
tokens/s
109
tokens/s
111
ms
pp2048+tg768
20177
tokens/s
109
tokens/s
111
ms
pp1024+tg1024
21826
tokens/s
109
tokens/s
56
ms
pp1280+tg3072
19385
tokens/s
106
tokens/s
75
ms
pp384+tg1152
18705
tokens/s
109
tokens/s
29
ms
pp64+tg1024
3821
tokens/s
111
tokens/s
26
ms
pp16+tg1536
1152
tokens/s
110
tokens/s
23
ms