TEST #3497 RESULTS
05/10/2026 - 8:17 PM
ACCELERATOR
MODEL
114
tokens/s
generation
131
ms
time to first token
10350
tokens/s
prompt
2083
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-1630 v3 @ 3.70GHz (haswell)
RAM
127.9GB
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
14361
tokens/s
98.3
tokens/s
99
ms
pp4096+tg256
9917
tokens/s
124
tokens/s
422
ms
pp2048+tg256
12772
tokens/s
121
tokens/s
169
ms
pp2048+tg768
12567
tokens/s
121
tokens/s
171
ms
pp1024+tg1024
12898
tokens/s
121
tokens/s
88
ms
pp1280+tg3072
13588
tokens/s
112
tokens/s
103
ms
pp384+tg1152
13764
tokens/s
107
tokens/s
41
ms
pp64+tg1024
2784
tokens/s
112
tokens/s
40
ms
pp16+tg1536
504
tokens/s
114
tokens/s
45
ms