TEST #1447 RESULTS
08/28/2025 - 3:35 PM
ACCELERATOR
MODEL
117
tokens/s
generation
349
ms
time to first token
3811
tokens/s
prompt
1084
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-7820X CPU @ 3.60GHz (skylake-avx512)
RAM
31.7GB
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
5089
tokens/s
125
tokens/s
209
ms
pp4096+tg256
3352
tokens/s
101
tokens/s
1.24
sec
pp2048+tg256
4254
tokens/s
116
tokens/s
494
ms
pp2048+tg768
4193
tokens/s
115
tokens/s
499
ms
pp1024+tg1024
4968
tokens/s
122
tokens/s
218
ms
pp1280+tg3072
4446
tokens/s
111
tokens/s
297
ms
pp384+tg1152
4958
tokens/s
122
tokens/s
88
ms
pp64+tg1024
2739
tokens/s
120
tokens/s
34
ms
pp16+tg1536
300
tokens/s
116
tokens/s
63
ms