TEST #3345 RESULTS
04/18/2026 - 7:38 AM
ACCELERATOR
MODEL
44.7
tokens/s
generation
1.30
sec
time to first token
986
tokens/s
prompt
324
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2673 v3 @ 2.40GHz (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
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1195
tokens/s
45.0
tokens/s
880
ms
pp4096+tg256
935
tokens/s
30.8
tokens/s
4.41
sec
pp2048+tg256
1089
tokens/s
41.2
tokens/s
1.91
sec
pp2048+tg768
1095
tokens/s
39.3
tokens/s
1.90
sec
pp1024+tg1024
1190
tokens/s
46.1
tokens/s
881
ms
pp1280+tg3072
1150
tokens/s
37.2
tokens/s
1.13
sec
pp384+tg1152
1229
tokens/s
52.1
tokens/s
330
ms
pp64+tg1024
905
tokens/s
56.9
tokens/s
86
ms
pp16+tg1536
90
tokens/s
54.0
tokens/s
195
ms