TEST #983 RESULTS
06/03/2025 - 3:46 AM
ACCELERATOR
MODEL
23.5
tokens/s
generation
4.54
sec
time to first token
594
tokens/s
prompt
146
LocalScore
HOW YOU STACK UP
Explore All ResultsGranite 3.3 8b Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-4590 CPU @ 3.30GHz (haswell)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
795
tokens/s
24.5
tokens/s
1.33
sec
pp4096+tg256
143
tokens/s
21.3
tokens/s
28.60
sec
pp2048+tg256
719
tokens/s
23.3
tokens/s
2.89
sec
pp2048+tg768
720
tokens/s
23.1
tokens/s
2.89
sec
pp1024+tg1024
781
tokens/s
24.0
tokens/s
1.35
sec
pp1280+tg3072
423
tokens/s
22.5
tokens/s
3.07
sec
pp384+tg1152
860
tokens/s
24.3
tokens/s
488
ms
pp64+tg1024
706
tokens/s
24.5
tokens/s
131
ms
pp16+tg1536
203
tokens/s
24.4
tokens/s
121
ms