TEST #1677 RESULTS
10/18/2025 - 6:50 PM
ACCELERATOR
MODEL
146
tokens/s
generation
318
ms
time to first token
4068
tokens/s
prompt
1232
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper 1920X 12-Core Processor
RAM
62.6GB
OS
Linux
Kernel Release
6.8.0-85-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #85-Ubuntu SMP PREEMPT_DYNAMIC Thu Sep 18 15:26:59 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5153
tokens/s
143
tokens/s
207
ms
pp4096+tg256
3742
tokens/s
103
tokens/s
1.11
sec
pp2048+tg256
4523
tokens/s
137
tokens/s
460
ms
pp2048+tg768
4561
tokens/s
131
tokens/s
456
ms
pp1024+tg1024
4828
tokens/s
151
tokens/s
219
ms
pp1280+tg3072
4540
tokens/s
124
tokens/s
289
ms
pp384+tg1152
5216
tokens/s
167
tokens/s
79
ms
pp64+tg1024
2866
tokens/s
182
tokens/s
27
ms
pp16+tg1536
1181
tokens/s
176
tokens/s
19
ms