TEST #4093 RESULTS
08/27/2026 - 7:24 PM
ACCELERATOR
MODEL
44.2
tokens/s
generation
6.53
sec
time to first token
218
tokens/s
prompt
114
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i7-11700K @ 3.60GHz (rocketlake)
RAM
31.8GB
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
230
tokens/s
45.5
tokens/s
4.48
sec
pp4096+tg256
195
tokens/s
35.2
tokens/s
21.01
sec
pp2048+tg256
202
tokens/s
42.5
tokens/s
10.18
sec
pp2048+tg768
199
tokens/s
42.8
tokens/s
10.33
sec
pp1024+tg1024
219
tokens/s
46.0
tokens/s
4.70
sec
pp1280+tg3072
214
tokens/s
40.0
tokens/s
6.00
sec
pp384+tg1152
230
tokens/s
47.6
tokens/s
1.69
sec
pp64+tg1024
245
tokens/s
49.6
tokens/s
280
ms
pp16+tg1536
230
tokens/s
48.9
tokens/s
89
ms