TEST #2159 RESULTS
01/06/2026 - 2:16 AM
ACCELERATOR
MODEL
118
tokens/s
generation
204
ms
time to first token
7480
tokens/s
prompt
1631
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.6GB
OS
Linux
Kernel Release
6.14.0-1015-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #15-Ubuntu SMP PREEMPT_DYNAMIC Tue Nov 25 18:02:16 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
9246
tokens/s
108
tokens/s
120
ms
pp4096+tg256
5062
tokens/s
107
tokens/s
819
ms
pp2048+tg256
7450
tokens/s
119
tokens/s
283
ms
pp2048+tg768
7513
tokens/s
121
tokens/s
280
ms
pp1024+tg1024
9456
tokens/s
130
tokens/s
115
ms
pp1280+tg3072
9070
tokens/s
118
tokens/s
148
ms
pp384+tg1152
12432
tokens/s
125
tokens/s
38
ms
pp64+tg1024
5422
tokens/s
123
tokens/s
19
ms
pp16+tg1536
1670
tokens/s
115
tokens/s
16
ms