TEST #955 RESULTS
05/30/2025 - 4:42 AM
ACCELERATOR
MODEL
142
tokens/s
generation
356
ms
time to first token
4094
tokens/s
prompt
1177
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
15.3GB
OS
Linux
Kernel Release
6.11.0-26-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #26~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Apr 17 19:20:47 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4817
tokens/s
152
tokens/s
219
ms
pp4096+tg256
3082
tokens/s
117
tokens/s
1.34
sec
pp2048+tg256
4074
tokens/s
137
tokens/s
510
ms
pp2048+tg768
4052
tokens/s
134
tokens/s
513
ms
pp1024+tg1024
4881
tokens/s
145
tokens/s
216
ms
pp1280+tg3072
4564
tokens/s
129
tokens/s
287
ms
pp384+tg1152
5725
tokens/s
151
tokens/s
74
ms
pp64+tg1024
4932
tokens/s
157
tokens/s
19
ms
pp16+tg1536
719
tokens/s
154
tokens/s
28
ms