TEST #168 RESULTS
03/30/2025 - 8:07 PM
ACCELERATOR
MODEL
101
tokens/s
generation
371
ms
time to first token
8328
tokens/s
prompt
1312
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 9800X3D 8-Core Processor
RAM
123.5GB
OS
Linux
Kernel Release
5.15.0-134-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #145-Ubuntu SMP Wed Feb 12 20:08:39 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
11403
tokens/s
116
tokens/s
98
ms
pp4096+tg256
7187
tokens/s
62.6
tokens/s
586
ms
pp2048+tg256
9477
tokens/s
91.7
tokens/s
227
ms
pp2048+tg768
9457
tokens/s
91.6
tokens/s
228
ms
pp1024+tg1024
11397
tokens/s
105
tokens/s
99
ms
pp1280+tg3072
10117
tokens/s
85.3
tokens/s
136
ms
pp384+tg1152
12294
tokens/s
113
tokens/s
40
ms
pp64+tg1024
3616
tokens/s
122
tokens/s
25
ms
pp16+tg1536
8
tokens/s
118
tokens/s
1.90
sec