TEST #1701 RESULTS
10/26/2025 - 3:14 AM
ACCELERATOR
MODEL
5.9
tokens/s
generation
3.36
sec
time to first token
399
tokens/s
prompt
89
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 PRO 8845HS w/ Radeon 780M Graphics (znver4)
RAM
45.8GB
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
476
tokens/s
6.3
tokens/s
2.32
sec
pp4096+tg256
373
tokens/s
1.7
tokens/s
11.58
sec
pp2048+tg256
444
tokens/s
4.1
tokens/s
4.89
sec
pp2048+tg768
445
tokens/s
3.7
tokens/s
4.88
sec
pp1024+tg1024
484
tokens/s
5.6
tokens/s
2.28
sec
pp1280+tg3072
465
tokens/s
3.2
tokens/s
2.94
sec
pp384+tg1152
510
tokens/s
8.1
tokens/s
836
ms
pp64+tg1024
206
tokens/s
11.4
tokens/s
360
ms
pp16+tg1536
193
tokens/s
9.2
tokens/s
125
ms