TEST #2903 RESULTS
02/26/2026 - 11:12 AM
ACCELERATOR
MODEL
158
tokens/s
generation
201
ms
time to first token
7437
tokens/s
prompt
1801
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.7GB
OS
Linux
Kernel Release
6.17.0-1008-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #8-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 21 17:56:56 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
8774
tokens/s
166
tokens/s
123
ms
pp4096+tg256
5215
tokens/s
132
tokens/s
793
ms
pp2048+tg256
7437
tokens/s
153
tokens/s
282
ms
pp2048+tg768
7708
tokens/s
149
tokens/s
272
ms
pp1024+tg1024
9187
tokens/s
160
tokens/s
117
ms
pp1280+tg3072
8987
tokens/s
147
tokens/s
149
ms
pp384+tg1152
12822
tokens/s
169
tokens/s
36
ms
pp64+tg1024
5220
tokens/s
173
tokens/s
18
ms
pp16+tg1536
1582
tokens/s
170
tokens/s
17
ms