TEST #1896 RESULTS
12/02/2025 - 3:53 PM
ACCELERATOR
MODEL
128
tokens/s
generation
185
ms
time to first token
7374
tokens/s
prompt
1720
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon Gold 5220R CPU @ 2.20GHz (cascadelake)
RAM
63.6GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
9711
tokens/s
126
tokens/s
113
ms
pp4096+tg256
6272
tokens/s
128
tokens/s
661
ms
pp2048+tg256
8213
tokens/s
129
tokens/s
257
ms
pp2048+tg768
8099
tokens/s
129
tokens/s
260
ms
pp1024+tg1024
9519
tokens/s
129
tokens/s
115
ms
pp1280+tg3072
8877
tokens/s
127
tokens/s
152
ms
pp384+tg1152
10957
tokens/s
126
tokens/s
44
ms
pp64+tg1024
4200
tokens/s
127
tokens/s
23
ms
pp16+tg1536
517
tokens/s
128
tokens/s
38
ms