TEST #346 RESULTS
04/06/2025 - 2:31 AM
ACCELERATOR
MODEL
55.0
tokens/s
generation
154
ms
time to first token
8761
tokens/s
prompt
1464
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
63.1GB
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
11270
tokens/s
53.2
tokens/s
123
ms
pp4096+tg256
9840
tokens/s
46.3
tokens/s
438
ms
pp2048+tg256
11788
tokens/s
60.2
tokens/s
200
ms
pp2048+tg768
11423
tokens/s
52.3
tokens/s
211
ms
pp1024+tg1024
11098
tokens/s
57.9
tokens/s
108
ms
pp1280+tg3072
11320
tokens/s
51.2
tokens/s
131
ms
pp384+tg1152
9298
tokens/s
58.5
tokens/s
78
ms
pp64+tg1024
2306
tokens/s
58.6
tokens/s
43
ms
pp16+tg1536
510
tokens/s
56.9
tokens/s
51
ms