TEST #4098 RESULTS
08/28/2026 - 5:44 PM
ACCELERATOR
MODEL
102
tokens/s
generation
1.82
sec
time to first token
772
tokens/s
prompt
351
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Genuine Intel CPU 0000%@ (sapphirerapids)
RAM
125.5GB
OS
Linux
Kernel Release
6.12.105+deb13-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.12.105-1 (2026-08-24)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
407
tokens/s
73.6
tokens/s
2.53
sec
pp4096+tg256
686
tokens/s
109
tokens/s
5.98
sec
pp2048+tg256
853
tokens/s
80.2
tokens/s
2.41
sec
pp2048+tg768
833
tokens/s
111
tokens/s
2.47
sec
pp1024+tg1024
944
tokens/s
101
tokens/s
1.09
sec
pp1280+tg3072
911
tokens/s
108
tokens/s
1.42
sec
pp384+tg1152
1010
tokens/s
98.2
tokens/s
388
ms
pp64+tg1024
834
tokens/s
127
tokens/s
84
ms
pp16+tg1536
474
tokens/s
111
tokens/s
40
ms