TEST #2766 RESULTS
02/21/2026 - 1:10 AM
ACCELERATOR
MODEL
214
tokens/s
generation
48
ms
time to first token
28446
tokens/s
prompt
5024
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 9 285K (arrowlake-s)
RAM
63.4GB
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
39002
tokens/s
188
tokens/s
32
ms
pp4096+tg256
25869
tokens/s
204
tokens/s
163
ms
pp2048+tg256
34344
tokens/s
218
tokens/s
64
ms
pp2048+tg768
34200
tokens/s
218
tokens/s
65
ms
pp1024+tg1024
39882
tokens/s
221
tokens/s
30
ms
pp1280+tg3072
36442
tokens/s
211
tokens/s
40
ms
pp384+tg1152
35781
tokens/s
222
tokens/s
15
ms
pp64+tg1024
8273
tokens/s
223
tokens/s
12
ms
pp16+tg1536
2218
tokens/s
222
tokens/s
12
ms