TEST #2753 RESULTS
02/18/2026 - 9:50 AM
ACCELERATOR
MODEL
19.2
tokens/s
generation
32.29
sec
time to first token
45
tokens/s
prompt
30
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-3570K CPU @ 3.40GHz (ivybridge)
RAM
15.7GB
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
45
tokens/s
22.1
tokens/s
22.70
sec
pp4096+tg256
38
tokens/s
16.1
tokens/s
108.29
sec
pp2048+tg256
46
tokens/s
17.0
tokens/s
44.96
sec
pp2048+tg768
37
tokens/s
13.9
tokens/s
55.20
sec
pp1024+tg1024
44
tokens/s
20.5
tokens/s
23.14
sec
pp1280+tg3072
48
tokens/s
18.3
tokens/s
26.83
sec
pp384+tg1152
49
tokens/s
18.0
tokens/s
7.84
sec
pp64+tg1024
52
tokens/s
23.4
tokens/s
1.26
sec
pp16+tg1536
50
tokens/s
23.3
tokens/s
361
ms