TEST #3892 RESULTS
07/12/2026 - 5:56 AM
ACCELERATOR
MODEL
9.0
tokens/s
generation
10.86
sec
time to first token
129
tokens/s
prompt
47
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 2600 Six-Core Processor
RAM
7.7GB
OS
Linux
Kernel Release
7.0.12-arch1-1
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Wed, 10 Jun 2026 08:58:02 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
137
tokens/s
10.9
tokens/s
7.57
sec
pp4096+tg256
113
tokens/s
3.8
tokens/s
36.60
sec
pp2048+tg256
126
tokens/s
6.5
tokens/s
16.38
sec
pp2048+tg768
126
tokens/s
6.0
tokens/s
16.38
sec
pp1024+tg1024
135
tokens/s
8.5
tokens/s
7.68
sec
pp1280+tg3072
133
tokens/s
5.3
tokens/s
9.73
sec
pp384+tg1152
141
tokens/s
11.6
tokens/s
2.78
sec
pp64+tg1024
142
tokens/s
15.5
tokens/s
486
ms
pp16+tg1536
103
tokens/s
13.2
tokens/s
188
ms