TEST #997 RESULTS
06/04/2025 - 7:09 PM
ACCELERATOR
MODEL
30.8
tokens/s
generation
6.63
sec
time to first token
236
tokens/s
prompt
103
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 4800H with Radeon Graphics (znver2)
RAM
15GB
OS
Linux
Kernel Release
6.14.9-300.fc42.x86_64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Thu May 29 14:27:53 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
240
tokens/s
32.8
tokens/s
4.30
sec
pp4096+tg256
170
tokens/s
26.4
tokens/s
24.09
sec
pp2048+tg256
208
tokens/s
30.2
tokens/s
9.87
sec
pp2048+tg768
209
tokens/s
29.6
tokens/s
9.83
sec
pp1024+tg1024
239
tokens/s
31.2
tokens/s
4.31
sec
pp1280+tg3072
233
tokens/s
28.6
tokens/s
5.53
sec
pp384+tg1152
273
tokens/s
32.5
tokens/s
1.43
sec
pp64+tg1024
296
tokens/s
33.4
tokens/s
249
ms
pp16+tg1536
255
tokens/s
32.8
tokens/s
91
ms