TEST #2252 RESULTS
01/12/2026 - 11:46 AM
ACCELERATOR
MODEL
129
tokens/s
generation
245
ms
time to first token
5750
tokens/s
prompt
1445
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
QEMU Virtual CPU version 2.5+
RAM
7.8GB
OS
Linux
Kernel Release
6.8.0-90-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #91-Ubuntu SMP PREEMPT_DYNAMIC Tue Nov 18 14:14:30 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
7246
tokens/s
127
tokens/s
149
ms
pp4096+tg256
4540
tokens/s
107
tokens/s
911
ms
pp2048+tg256
5976
tokens/s
122
tokens/s
351
ms
pp2048+tg768
5954
tokens/s
124
tokens/s
352
ms
pp1024+tg1024
7194
tokens/s
127
tokens/s
150
ms
pp1280+tg3072
6787
tokens/s
117
tokens/s
197
ms
pp384+tg1152
8367
tokens/s
134
tokens/s
53
ms
pp64+tg1024
4745
tokens/s
137
tokens/s
21
ms
pp16+tg1536
941
tokens/s
164
tokens/s
24
ms