TEST #1574 RESULTS
09/30/2025 - 6:59 AM
ACCELERATOR
MODEL
41.5
tokens/s
generation
5.77
sec
time to first token
252
tokens/s
prompt
122
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i7-13800H (alderlake)
RAM
61.2GB
OS
Linux
Kernel Release
6.14.0-32-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #32-Ubuntu SMP PREEMPT_DYNAMIC Fri Aug 29 14:21:26 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
261
tokens/s
43.2
tokens/s
3.94
sec
pp4096+tg256
205
tokens/s
35.4
tokens/s
20.04
sec
pp2048+tg256
239
tokens/s
41.7
tokens/s
8.61
sec
pp2048+tg768
239
tokens/s
41.1
tokens/s
8.60
sec
pp1024+tg1024
260
tokens/s
42.0
tokens/s
3.96
sec
pp1280+tg3072
255
tokens/s
38.3
tokens/s
5.05
sec
pp384+tg1152
280
tokens/s
43.8
tokens/s
1.40
sec
pp64+tg1024
281
tokens/s
44.6
tokens/s
249
ms
pp16+tg1536
246
tokens/s
43.5
tokens/s
87
ms