TEST #1474 RESULTS
09/04/2025 - 1:03 AM
ACCELERATOR
MODEL
1.2
tokens/s
generation
747.08
sec
time to first token
2
tokens/s
prompt
1
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core2 Duo CPU E8135 @ 2.66GHz
RAM
7.5GB
OS
Linux
Kernel Release
6.8.0-71-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #71-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 22 16:52:38 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2
tokens/s
1.4
tokens/s
476.81
sec
pp4096+tg256
2
tokens/s
0.9
tokens/s
2685.85
sec
pp2048+tg256
2
tokens/s
1.1
tokens/s
1125.20
sec
pp2048+tg768
2
tokens/s
1.1
tokens/s
1139.46
sec
pp1024+tg1024
2
tokens/s
1.2
tokens/s
477.46
sec
pp1280+tg3072
2
tokens/s
1.1
tokens/s
614.82
sec
pp384+tg1152
2
tokens/s
1.3
tokens/s
170.90
sec
pp64+tg1024
3
tokens/s
1.4
tokens/s
26.11
sec
pp16+tg1536
2
tokens/s
1.4
tokens/s
7.09
sec