TEST #1537 RESULTS
09/22/2025 - 3:17 PM
ACCELERATOR
MODEL
90.1
tokens/s
generation
213
ms
time to first token
6763
tokens/s
prompt
1419
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 8945HS w/ Radeon 780M Graphics (znver4)
RAM
30.6GB
OS
Linux
Kernel Release
6.16.8-2-cachyos
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Fri, 19 Sep 2025 15:36:25 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
9222
tokens/s
100
tokens/s
123
ms
pp4096+tg256
5417
tokens/s
73.3
tokens/s
773
ms
pp2048+tg256
7168
tokens/s
87.0
tokens/s
300
ms
pp2048+tg768
6882
tokens/s
83.4
tokens/s
312
ms
pp1024+tg1024
9374
tokens/s
92.0
tokens/s
125
ms
pp1280+tg3072
7697
tokens/s
78.7
tokens/s
179
ms
pp384+tg1152
11218
tokens/s
96.8
tokens/s
49
ms
pp64+tg1024
2768
tokens/s
101
tokens/s
38
ms
pp16+tg1536
1122
tokens/s
98.7
tokens/s
24
ms