TEST #851 RESULTS
05/10/2025 - 2:34 PM
ACCELERATOR
MODEL
16.7
tokens/s
generation
5.53
sec
time to first token
232
tokens/s
prompt
89
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2667 v3 @ 3.20GHz (haswell)
RAM
62.8GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
297
tokens/s
18.0
tokens/s
3.50
sec
pp4096+tg256
238
tokens/s
15.4
tokens/s
17.30
sec
pp2048+tg256
265
tokens/s
17.4
tokens/s
7.81
sec
pp2048+tg768
242
tokens/s
16.5
tokens/s
8.51
sec
pp1024+tg1024
263
tokens/s
18.3
tokens/s
3.95
sec
pp1280+tg3072
209
tokens/s
17.0
tokens/s
6.18
sec
pp384+tg1152
199
tokens/s
16.1
tokens/s
1.99
sec
pp64+tg1024
212
tokens/s
16.1
tokens/s
350
ms
pp16+tg1536
159
tokens/s
15.5
tokens/s
149
ms