TEST #689 RESULTS
04/11/2025 - 11:00 AM
ACCELERATOR
MODEL
36.7
tokens/s
generation
2.59
sec
time to first token
1540
tokens/s
prompt
279
LocalScore
HOW YOU STACK UP
Explore All ResultsGemma 3 12b It - Q3_K - Large
SYSTEM
CPU
13th Gen Intel Core i7-13700K (alderlake)
RAM
63.8GB
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
2373
tokens/s
40.3
tokens/s
461
ms
pp4096+tg256
254
tokens/s
15.2
tokens/s
16.22
sec
pp2048+tg256
2136
tokens/s
40.0
tokens/s
985
ms
pp2048+tg768
2111
tokens/s
41.4
tokens/s
996
ms
pp1024+tg1024
2411
tokens/s
42.4
tokens/s
449
ms
pp1280+tg3072
333
tokens/s
26.9
tokens/s
3.87
sec
pp384+tg1152
2381
tokens/s
40.8
tokens/s
185
ms
pp64+tg1024
1345
tokens/s
41.3
tokens/s
73
ms
pp16+tg1536
516
tokens/s
42.1
tokens/s
55
ms