TEST #3872 RESULTS
07/07/2026 - 4:47 PM
ACCELERATOR
MODEL
34.3
tokens/s
generation
13.36
sec
time to first token
109
tokens/s
prompt
65
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 2600 Six-Core Processor
RAM
31.3GB
OS
Linux
Kernel Release
6.12.90+deb13.1-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.12.90-2 (2026-05-27)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
113
tokens/s
37.5
tokens/s
9.09
sec
pp4096+tg256
89
tokens/s
26.8
tokens/s
46.22
sec
pp2048+tg256
102
tokens/s
32.8
tokens/s
20.02
sec
pp2048+tg768
103
tokens/s
31.8
tokens/s
19.97
sec
pp1024+tg1024
111
tokens/s
34.9
tokens/s
9.25
sec
pp1280+tg3072
109
tokens/s
30.1
tokens/s
11.74
sec
pp384+tg1152
119
tokens/s
37.3
tokens/s
3.26
sec
pp64+tg1024
124
tokens/s
39.3
tokens/s
541
ms
pp16+tg1536
115
tokens/s
38.1
tokens/s
163
ms