TEST #3874 RESULTS
07/07/2026 - 9:43 PM
ACCELERATOR
MODEL
124
tokens/s
generation
421
ms
time to first token
3175
tokens/s
prompt
977
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-6700K CPU @ 4.00GHz (skylake)
RAM
60.5GB
OS
Linux
Kernel Release
6.14.0-37-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #37-Ubuntu SMP PREEMPT_DYNAMIC Fri Nov 14 22:10:32 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4005
tokens/s
131
tokens/s
263
ms
pp4096+tg256
2803
tokens/s
86.5
tokens/s
1.47
sec
pp2048+tg256
3636
tokens/s
110
tokens/s
572
ms
pp2048+tg768
3105
tokens/s
109
tokens/s
669
ms
pp1024+tg1024
3726
tokens/s
127
tokens/s
282
ms
pp1280+tg3072
3413
tokens/s
105
tokens/s
383
ms
pp384+tg1152
3984
tokens/s
144
tokens/s
103
ms
pp64+tg1024
2819
tokens/s
155
tokens/s
28
ms
pp16+tg1536
1082
tokens/s
149
tokens/s
20
ms