TEST #338 RESULTS
04/06/2025 - 1:35 AM
ACCELERATOR
MODEL
52.5
tokens/s
generation
205
ms
time to first token
6597
tokens/s
prompt
1190
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
31.6GB
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
8644
tokens/s
55.5
tokens/s
138
ms
pp4096+tg256
6475
tokens/s
51.4
tokens/s
653
ms
pp2048+tg256
8101
tokens/s
53.9
tokens/s
273
ms
pp2048+tg768
8064
tokens/s
52.2
tokens/s
273
ms
pp1024+tg1024
8900
tokens/s
52.7
tokens/s
144
ms
pp1280+tg3072
8058
tokens/s
52.3
tokens/s
177
ms
pp384+tg1152
8903
tokens/s
50.7
tokens/s
86
ms
pp64+tg1024
1694
tokens/s
51.8
tokens/s
56
ms
pp16+tg1536
534
tokens/s
51.7
tokens/s
49
ms