TEST #3201 RESULTS
03/28/2026 - 1:45 PM
ACCELERATOR
MODEL
137
tokens/s
generation
95
ms
time to first token
13908
tokens/s
prompt
2716
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
61.9GB
OS
Linux
Kernel Release
6.18.7-76061807-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202601231045~1769703228~24.04~cb87b5b SMP PREEMPT_DYNAMIC Thu J
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
18578
tokens/s
158
tokens/s
61
ms
pp4096+tg256
13061
tokens/s
105
tokens/s
323
ms
pp2048+tg256
16440
tokens/s
135
tokens/s
132
ms
pp2048+tg768
16471
tokens/s
122
tokens/s
132
ms
pp1024+tg1024
18254
tokens/s
143
tokens/s
63
ms
pp1280+tg3072
17125
tokens/s
119
tokens/s
82
ms
pp384+tg1152
18518
tokens/s
150
tokens/s
29
ms
pp64+tg1024
5009
tokens/s
152
tokens/s
19
ms
pp16+tg1536
1718
tokens/s
151
tokens/s
15
ms