TEST #1278 RESULTS
07/31/2025 - 3:12 PM
ACCELERATOR
MODEL
43.9
tokens/s
generation
5.08
sec
time to first token
279
tokens/s
prompt
134
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600G with Radeon Graphics (znver3)
RAM
46.9GB
OS
Linux
Kernel Release
6.11.0-29-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #29~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Jun 26 14:16:59 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
311
tokens/s
48.6
tokens/s
3.31
sec
pp4096+tg256
234
tokens/s
37.1
tokens/s
17.51
sec
pp2048+tg256
271
tokens/s
42.9
tokens/s
7.57
sec
pp2048+tg768
269
tokens/s
42.6
tokens/s
7.63
sec
pp1024+tg1024
265
tokens/s
44.7
tokens/s
3.89
sec
pp1280+tg3072
299
tokens/s
40.4
tokens/s
4.30
sec
pp384+tg1152
343
tokens/s
45.3
tokens/s
1.14
sec
pp64+tg1024
312
tokens/s
46.3
tokens/s
224
ms
pp16+tg1536
206
tokens/s
47.3
tokens/s
99
ms