TEST #1071 RESULTS
06/18/2025 - 2:58 PM
ACCELERATOR
MODEL
11.8
tokens/s
generation
1.96
sec
time to first token
666
tokens/s
prompt
159
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 8845HS w/ Radeon 780M Graphics (znver4)
RAM
27.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
836
tokens/s
12.8
tokens/s
1.30
sec
pp4096+tg256
639
tokens/s
3.3
tokens/s
6.71
sec
pp2048+tg256
670
tokens/s
8.4
tokens/s
3.19
sec
pp2048+tg768
789
tokens/s
7.7
tokens/s
2.73
sec
pp1024+tg1024
872
tokens/s
11.2
tokens/s
1.25
sec
pp1280+tg3072
830
tokens/s
6.5
tokens/s
1.63
sec
pp384+tg1152
809
tokens/s
15.9
tokens/s
518
ms
pp64+tg1024
319
tokens/s
22.0
tokens/s
227
ms
pp16+tg1536
225
tokens/s
18.3
tokens/s
102
ms