TEST #366 RESULTS
04/06/2025 - 1:15 PM
ACCELERATOR
MODEL
126
tokens/s
generation
352
ms
time to first token
4078
tokens/s
prompt
1134
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i7-12700H (alderlake)
RAM
31.7GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4973
tokens/s
132
tokens/s
214
ms
pp4096+tg256
3115
tokens/s
107
tokens/s
1.32
sec
pp2048+tg256
4147
tokens/s
123
tokens/s
502
ms
pp2048+tg768
4140
tokens/s
121
tokens/s
503
ms
pp1024+tg1024
4960
tokens/s
129
tokens/s
214
ms
pp1280+tg3072
4661
tokens/s
117
tokens/s
282
ms
pp384+tg1152
5729
tokens/s
133
tokens/s
75
ms
pp64+tg1024
4286
tokens/s
135
tokens/s
22
ms
pp16+tg1536
693
tokens/s
134
tokens/s
30
ms