TEST #2553 RESULTS
02/07/2026 - 3:45 AM
ACCELERATOR
MODEL
22.9
tokens/s
generation
7.21
sec
time to first token
209
tokens/s
prompt
87
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-10700K CPU @ 3.80GHz (skylake)
RAM
15.9GB
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
215
tokens/s
23.8
tokens/s
4.80
sec
pp4096+tg256
158
tokens/s
20.2
tokens/s
25.91
sec
pp2048+tg256
192
tokens/s
22.5
tokens/s
10.72
sec
pp2048+tg768
195
tokens/s
22.1
tokens/s
10.53
sec
pp1024+tg1024
211
tokens/s
23.3
tokens/s
4.89
sec
pp1280+tg3072
215
tokens/s
21.5
tokens/s
6.00
sec
pp384+tg1152
247
tokens/s
24.1
tokens/s
1.59
sec
pp64+tg1024
244
tokens/s
24.7
tokens/s
304
ms
pp16+tg1536
201
tokens/s
24.3
tokens/s
119
ms