TEST #2943 RESULTS
03/02/2026 - 3:20 AM
ACCELERATOR
MODEL
12.0
tokens/s
generation
2.05
sec
time to first token
632
tokens/s
prompt
155
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 8945HS w/ Radeon 780M Graphics (znver4)
RAM
29.8GB
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
820
tokens/s
13.1
tokens/s
1.32
sec
pp4096+tg256
598
tokens/s
3.7
tokens/s
7.12
sec
pp2048+tg256
720
tokens/s
8.7
tokens/s
2.97
sec
pp2048+tg768
722
tokens/s
7.9
tokens/s
2.96
sec
pp1024+tg1024
759
tokens/s
11.5
tokens/s
1.42
sec
pp1280+tg3072
758
tokens/s
6.6
tokens/s
1.78
sec
pp384+tg1152
777
tokens/s
16.2
tokens/s
537
ms
pp64+tg1024
299
tokens/s
22.3
tokens/s
240
ms
pp16+tg1536
235
tokens/s
18.5
tokens/s
98
ms